Equity Valuation of University-based

Spinoffs: A Literature Review and

Empirical Study

Arif Muhammed Mirza

Ole Rønning

NTNU School of Entrepreneurship

Submission date: June 2015
Supervisor: Roger Sørheim, IØT
Co-supervisor: Marius Tuft Mathisen, IØT

Norwegian University of Science and Technology

Department of Industrial Economics and Technology Management
Problem Description
This study examines valuation of early stage USOs using traditional valuation methods as well as
the feasibility of a resource-based valuation framework.
Preface
This study is the master thesis of Arif Mirza and Ole Rønning, two graduate M.Sc. students at the
NTNU School of Entrepreneurship. The motivation behind the chosen research topic is grounded
in both authors’ background in entrepreneurship, and more precisely their personal experience
with the difficulties associated with valuing young startups. Further on, the focus on university
spinoffs (USOs) was a natural development due to the authors residing in the technological capital
of Norway, Trondheim, which is the originating region of many of Norway’s USOs.

The purpose of this research is to shed light on the limitations associated with today’s valuation
methods used in the context of USOs, as well as establish the initial theoretical foundation for an
alternative valuation approach that might be better for this type of firms. The research is presented
in two separate articles. The work has provided the authors with deep knowledge about valuing
new ventures, and the phenomena of university spinoffs. Further on, the empirical analysis
provided the authors an opportunity to expand on their limited statistical knowledge. Through this
the authors have learned statistical tools we are certain will be useful as we venture into our
business careers. As a final remark, it has become evident to the authors that today’s valuation
methods have significant shortcomings when used in the context of USOs, and that the findings in
this study do indeed fill a gap in the literature on both valuation and university spinoffs.

The authors want to thank Ph.D. candidate Marius Tuft Mathisen, from the Norwegian University
of Science and Technology. His enthusiastic attitude towards the topic of research has been an
important motivator throughout our work, and his inputs and mentoring have been greatly
appreciated. We also wish to thank Professor Roger Sørheim and Øystein Widding for providing
valuable feedback and advice.

The Authors

Trondheim, June 10, 2015

Contact Information:
Arif Mirza
Ole Rønning
Summary
During the last three decades, there has been a significant growth in the establishment of university spin-
offs (USOs), and they have become an important part of the economic landscape. These are companies
where it is generally agreed upon that financing is essential in their early stages, yet who face difficulties
in raising the necessary capital. This is believed to be due to long development times, and large amounts of
uncertainty which makes these companies difficult to value. This master thesis is divided into two articles,
where the first examines the applicability of traditional valuation methods on USOs, and the second seeks
to investigate the feasibility of a valuation framework based on resource inputs, rather than financial
outputs. Together, these two articles aim to lay the foundation on which future research can build to
develop better valuation methods, and through this make it easier for investors to assess early stage
university spin-offs and close the financing gap these companies face.
Through article 1, the authors investigate, with a basis in a two dimensional theoretical framework based
on resource-based theory and risk and uncertainty, how unique resource characteristics in these firms
affect the perceived uncertainty seen from the viewpoint of an investor, and subsequently how this
uncertainty affects the applicability of traditional valuation methods from corporate finance. It is found
that the early-stage, radicalness and significant technical advances of the technologies many of these
companies are founded on lead to high fundamental uncertainty, which renders the three most common
valuation methods cash flow discounting, balance sheet and income statement inapplicable to early stage
USOs. This has important implications for investors, who by blindly using these methods on USOs can
make bad investment decisions, or forego valuable investment opportunities.
In light of the findings in article 1, article 2 investigates the feasibility of a resource-based valuation
framework, with the underlying argument that when it is difficult to value a company with a basis in
financial outputs used in traditional corporate finance, a method based on inputs identified through theories
from strategic management that help predict and explain superior firm performance may prove more
satisfactory. Followingly, a conceptual framework with a basis in resource-based theory is developed. To
test the framework, six hypotheses about the relationship between initial resource endowments in USOs,
and their long term equity value are formulated, and empirically tested through regression on a sample of
63 Norwegian USOs. It is found that, ceteris paribus, higher degrees of heterogeneity in the founding team,
better quality parent universities as well as more filed and published patents in the early stages of USO are
all positively related to the firms long term equity value. With a basis in this it is argued that USOs who
exhibit these resource characteristics in their early stages should be valued higher than those that do not.
This has implications for investors, who should take into consideration the particular resources a firm
holds in their value estimates, and for entrepreneurs who should seek to assemble specific resources to
increase the future value of their firm. Although the development of a complete valuation model was
outside of the articles scope, the conceptual valuation framework developed proves that a resource-based
valuation framework has potential, and future research efforts should be directed towards this topic.
!
Sammendrag
Oppstartsbedrifter med opphav fra universitetsforskning(USOs) har de siste årene vokst betydelig i antall,
og blitt en viktig del av det økonomiske landskapet. Dette er bedrifter hvor det generelt er enighet i at
kapital er spesielt viktig i tidligfase, men som møter store utfordringer i å innhente den nødvendige
kapitalen. Dette er til dels grunnet lange utviklingstider, og vanskeligheter for investorer i å vurdere
verdiene til disse selskapene. Denne masteroppgaven er delt i to artikler, hvor artikkel 1 undersøker
hvorvidt eksisterende verdsettelsesmetoder i en tilfredsstillende grad kan brukes til å finne verdien av
tidligfase USOs. Artikkel 2 undersøker muligheten for å utvikle et verdsettelsesrammeverk basert på
ressurser i selskapet. Sammen søker disse to artiklene å fylle forskningsmangelen på verdsettelse av disse
spesielle selskapene, og danne grunnlaget for utvikling av mer robuste verdsettelsesmetoder som letter
verdivurderingen, og videre kapitaltilgangen fra investorer.
Gjennom artikkel 1 søker forfatterne, med utgangspunkt i et todimensjonalt teoretiskrammeverk bygd på
ressursbasert teori og risiko og usikkerhet, å identifisere hva som gjør disse selskapene unike, hvordan
disse unike karakteristikkene påvirker usikkerheten knyttet til USOs sett fra en investorsståsted, og til slutt
hvordan dette igjen påvirker muligheten for å bruke tradisjonelle verdsettelsesmetoder. For å svare på disse
spørsmålene gjennomføres det et litteratursøk på både verdsetting og ressurser i USOs. Resultatene viser
at spesielt tre karakteristikker ved de teknologiske ressursene i tidligfase USOs, nemlig teknologiens
tidlige utviklingsstadium, dens radikale natur samt signifikante tekniske fremskritt, øker den fundamentale
usikkerheten i selskapene, noe som gjør det umulig å på en pålitelig måte estimere finansielle data
tradisjonelle verdsettelsesmetoder er avhengig av. Dette fører til at de har begrenset anvendbarhet på disse
selskapene. Dette har viktige implikasjoner for investorer, som ved å blindt bruke disse metodene kan
gjennomføre dårlige investeringer, eller gå glipp av gode investeringsobjekt.
I lys av funnene i artikkel 1, undersøker artikkel 2 muligheten for å utvikle et verdsettelsesrammeverk
basert på ressurser i selskapet i stedet for finansielle data, med den underliggende logikken at når det er
vanskelig å estimere fremtidig ytelse kan muligens en metode med fundament i teorier som forklarer og
spår ytelse føre til bedre resultater. Et konseptuelt verdsettelsesrammeverk for sammenhengen mellom
ressurser og selskapsverdi utvikles. For å teste rammeverket utvikles seks hypoteser med utgangspunkt i
ressurs-ytelsessammenhenger i USOs påvist i tidligere forskning. Disse hypotesene testes gjennom
regresjon på et datasett bestående av 63 norske USOs. Resultatene indikerer at USOs som i tidligfase har
mer heterogene oppstartsteam, som har opphav i universiteter av høyere kvalitet, og som har flere
publiserte og søkte patenter, har, gitt at alt annet holds konstant, høyere verdi i lengden. Det argumenteres
følgelig for at USOs som har disse ressurskarateristikkene i tidligfase bør verdsettes høyere enn selskaper
som ikke har de. Dette har implikasjoner for investorer, som i sine verdivurderinger kan bihensyna
ressurser i selskapene, og for grundere som bør søke å sample visse ressurser for å øke verdien på
selskapet sitt. Selv om utviklingen av et komplett verdsettelsesrammeverk er utenfor omfanget til
artikkelen, og resultatene kun indikativen, tyder de på at verdsetting basert på ressurser kan være
hensiktsmessig. For å verifisere dette er det nødvendig med betydelig fremtidig forskningsfokus på temaet.
!
 Examining Equity Valuation of University Spin-
off Companies: A Literature Review
Arif M. Mirza, Ole Rønning

ABSTRACT
University spinoffs have in the last three decades grown considerably in numbers, and have become an
important part of the economic landscape. This study investigates how unique characteristics of initial
resource endowments of these firms affect the applicability of traditional valuation methods from
corporate finance. To answer the research question, a two-dimensional theoretical framework consisting of
risk and uncertainty and resource-based theory is used. As an empirical foundation, literature reviews on
both valuation and university spinoffs are conducted. Our findings show that the three most common
groups of valuation methods, i.e. cash flow discounting, balance sheet and income statement, are not
applicable to early-stage USOs. This is primarily due to the technologies these companies are founded on
being early-stage, radical and providing significant technical advances, which leads to high fundamental
uncertainty. This uncertainty makes it difficult to reliably estimate financial variables on which traditional
valuation methods are dependent. The findings have particular implications for investors, who by blindly
using traditional valuation methods can overlook interesting prospects or make poor investment decisions.

average start-up (Shane, 2004; Clarysse and

Introduction Bruneel, 2007). Despite the need for early-stage
financing, USOs face an impediment in razing
There has been a notable increase in technology
the necessary capital due to high levels of
transfer from universities to the industry in the
uncertainty associated with these ventures
last decades (Rothaermel et al., 2007). From
(Widding et al., 2009; Shane, 2004).
1980 to 1993, American academic institutions
saw an increase of 444 percent in the “Finance is a catalyst of this
establishment of university spinoff companies wealth creation yet access to
(USOs) (Pressman, 2002). The US Bayh-Dole venture capital is a major
Act of 1980 and its European counterparts, a rise impediment faced by these
in venture capital, a larger pool and thus mobility companies”
of scientists and engineers, and important
– Wright et al. (2006)
technological breakthroughs in computing are
some of the factors attributed to this growth. Not An important step in obtaining financing for a
surprisingly, these significant developments in new venture is determining its value. Tyebjee
establishment of university spinoffs, and their and Bruno (1984) maintain that establishing the
potential economic impacts, have attracted an price of venture capital is the heart of any
increasing attention of research towards the negotiation between the founders of the venture
phenomenon both in the United States and in and potential investors. To determine the price,
Europe (Rothaermel et al., 2007). valuation becomes essential, and common
valuation methods have traditionally been
Some of this research has been on the topic of
dependent on future cash flows, assets or
financing of early-stage USOs, and more
valuations of comparative firms (Seppä and
specifically, the lack thereof. The majority of
Laamanen, 2001). A fundamental assumption
startups require access to financing to develop
underlying these methods is that there is an
their products and services, and obtain growth.
efficient capital market for ownership in the firm
However, researchers have argued that
(Timmons and Spinelli, 2004), i.e. that
financing, and specifically early-stage financing,
informational asymmetries are uncommon
might be more crucial for USOs compared to the
2 Arif M. Mirza, Ole Rønning

(Fama, 1991) and that asset prices therefore increasingly difficult to estimate under higher
reflect all publicly known information. For uncertainty.
publically traded firms this assumption may
This is what motivates this study. Blindly using
hold, but for new ventures with a lack of
existing valuation methods for the lack of better
liquidity of ownership, short operating history
approaches might yield significantly misleading
and limited/private accounting information,
valuations, leading investors to make bad
informational asymmetries between the investor
investment decisions or generally refrain from
and firm can be large, and it does not. Timmons
investing in early-stage USOs to avoid the risk of
and Spinelli (2004) argue that it is this
inadequately assessing their investment projects.
inefficiency in the venture capital market that
A better understanding of how to value these
makes the most common valuation methods
firms might help solve this problem.
prone to error when used on early-stage
Consequently we raise the following research
companies, because estimating cash flows and
question:
discount rates becomes difficult. In agreement
with Timmons and Spinelli (2004), Damodaran RQ: How do characteristics of
(2009) points out that young companies are initial resource endowments in
problematic to value due to operating losses or USOs affect the suitability of using
small revenues, multiple claims on equity, the traditional valuation methods?
fact that they are illiquid investments and that it
In order to answer the research question, we
is difficult to account for the risk of the company
draw upon resource-based theory as well as the
not surviving.
theory of risk and uncertainty. The logic being
The factors leading to inefficiency highlighted that the former helps us to identify the firm
by Timmons and Spinelli (2004) are especially characteristics that make these companies
prevalent in USOs, whom lately have become an unique, and the latter to assess how these
important part of the economic landscape characteristics affect the inherent difficulties in
through the creation of jobs and encouraging assessing the risk in early-stage USOs, and
local economic development (Shane, 2004; subsequently the difficulties in using traditional
Bonardo et al., 2011). This is a type of company valuation methods. In the next part we introduce
where it is generally agreed upon that the the theoretical framework underlying this study
financial needs are greater than those of other before we present the research methodology
firms and that capital is essential for growth employed. In the subsequent literature review we
(Cohen, 2006; Neck et al., 2004; Sætre et al., present relevant literature on both traditional
2006; Clarysse and Bruneel, 2007), yet Shane valuation methods as well as resource
(2004) states that they have difficulty in raising characteristics of early stage USOs. Finally we
capital in their early stages. This may partially be discuss how these initial resource characteristics
attributed to problems in valuing these affect the applicability of traditional valuation
companies. As Keeley et al. (1996) writes, methods, and what implications this has for
valuation is a central part in deciding whether investors and entrepreneurs.
one should invest in a company or not, but with
indications that these firms take on average 10 Definition of Early-Stage University
years before reaching the market (Rasmussen et Based Spinoffs
al., 2013; Lawton Smith and Ho, 2006), liquidity
of ownership is a major issue. With many of In the literature on university spinoffs there have
these companies being founded on research with been nearly as many names to the phenomenon
nothing more than a proof of concept (Shane, as there have been researchers studying it.
2004) it takes years before operating history is Authors have for example referred to these
available. Further on, advanced and complex companies as university startups (Criaco et al.,
technologies increase informational asymmetries 2013; Powers and McDougall, 2005) or
between the USO and possible investors (Shane, university-based startups (Marion et al., 2012),
2004). As Rasmussen et al. (2013) points out, as university spinoffs (Bigliardi et al., 2013;
USOs face significant uncertainty in their early Wright et al., 2012), new technology-based firms
stages due to unproven technology, uncertainty (Fuller and Rothaermel, 2012), research-based
about market need and organizational scientific organizations (Mustar et al., 2006),
uncertainty. Valuing such a company obviously academic spinoffs (Clarysse et al., 2007; Festel,
poses new challenges, since parameters that 2013) and academic startups (Colombo and Piva,
traditionally have been emphasized are cash 2012). Defining the phenomenon has been
flows and discount rates, which become equally inconsistent. Some definitions restrict the
 Examining Equity Valuation of University Spin-off Companies: A literature review 3

term to those companies where the intellectual Damodaran, 2009), and to investigate whether
property is transferred from the university to the the valuation of early-stage USOs indeed is a
new venture (Shane, 2004; Lockett et al., 2005), special case, it becomes necessary to understand
others also include companies founded on what makes these firms unique. To investigate
knowledge obtained at the parent university this the authors use resource-based theory as the
(Rappert et al., 1999), such as consultancy theoretical context to identify the unique
companies. resource characteristics of these firms. As Miller
(2003) writes, resources are the ‘raw material’ of
In this study, university spinoffs are defined as
business strategy, and therefore provide the basis
companies that involve transfer of intellectual
on which firms can distinguish themselves from
property rights (IPR) to the new venture, and that
others. However, simply identifying these
are either developed by faculty members based
differences will not be adequate to assess the
on their own research, or created specifically to
applicability of common valuation methods on
capitalize on academic research. This definition
early stage USOs. The underlying factor making
is keeping with the literature (Shane, 2004;
valuation of new ventures difficult is uncertainty,
Colombo et al., 2010; Lockett et al., 2005), and
which in the context of valuation becomes
is chosen because firms that involve transfer of
important to clearly separate from risk. To
IPR are more likely to have larger financial
understand this, we use the notion of risk and
needs, to seek funding in their early stages, and
uncertainty as the second dimension in the
followingly are more impacted by issues related
theoretical framework.
to valuation, than for example consultancy
companies founded on implicit knowledge
developed at the parent university. Further on, Risk, Uncertainty and Informational
we are interested in investigating early stage Asymmetries
valuation, and therefore focus on early-stage
Knight (1921) defines the concepts of risk and
USOs. Vohora et al. (2002) finds that USOs
uncertainty as two distinct variables based on
develop in five distinct phases: the research
three categories of probabilities: A priori
phase, opportunity framing phase, pre-
probability, statistical probability and estimates.
organization phase, re-orientation phase and
A priori probabilities are derived from absolutely
finally sustainable returns phase. The research
homogenous classification of instances, for
phase is where the intellectual property is
example as the case in a dice throw. Statistical
created, and is prior to any commercialization.
probability on the other hand is obtained through
The opportunity framing phase includes
analysis: “it is an empirical evaluation of the
screening of the technology with regards to
frequency of association between predicates…”
validity and performance, and research to prove
(Knight, 1921). Lastly, estimates are a group of
that the technology works outside the laboratory.
probabilities for which there is no valid basis of
Vohora et al. (2002) writes that a fundamental
any kind for classifying instances. It is this last
problem at this stage is that there is a mismatch
category of probabilities that Knight defines as
between what the university offers, and what
uncertainty, while the former two are defined as
venture capitalists want. The new ventures have
risk.
little proof of concept, no target market or
commercial management, and generally lacking “The practical difference between
resources to further develop the opportunity. It is the two categories, risk and
followingly in this stage that the financing gap uncertainty, is that in the former the
(Shane, 2004) starts to develop, and where the distribution of the outcome in a
problems of valuation become relevant. We group of instances is known (either
therefore focus on USOs in this stage of through calculation a priori or
development, and define them as early-stage from statistics of past experience),
USOs. while in the case of uncertainty this
is not true, the reason being in
Theoretical Framework general that it is impossible to form
a group of instances, because the
The topic of valuation is traditionally a field situation dealt with is in a high
within corporate finance, and most valuation degree unique.”
methods are, to some extent, based on
– Knight (1921)
accounting information. Valuing any startup
using these common methods is an inherently Dequech (2000) further distinguishes between
difficult task (Timmons and Spinelli, 2004; fundamental uncertainty and ambiguity. What
4 Arif M. Mirza, Ole Rønning

Dequech defines as fundamental uncertainty is in relationship between risk, ambiguity,

accordance with Knight’s definition of informational asymmetries, and fundamental
uncertainty. However, he goes on to argue that uncertainty, where an upwards movement on the
one has to distinguish between fundamental Y-axis denotes increasing uncertainty, while a
uncertainty and ambiguity, where he defines downwards movement indicates increasing risk.
ambiguity in accordance with Camerer and
Weber (1992):
“Ambiguity is uncertainty about
probability, created by missing
information that is relevant and
could be known. Ambiguity is
therefore not uncertainty about the
possible outcomes, but rather lack
of certainty about probabilities of
outcomes being 1, 0 or something
in between. Information is hidden,
rather than non-existent at the
moment of decision”
– Dequech (2000)
Ambiguity therefore, can lead to situations of
asymmetric information.Asymmetric
information denotes a situation in which one
contracting party has superior information
compared to another, and has consequences Figure 1. The relationship between risk, fundamental
uncertainty, ambiguity and informational asymmetries.
whenever a “principal”, who must contract with
an “agent” to accomplish some purpose, cannot The distinction between risk and uncertainty, as
be sure how well the agent will perform because well as informational asymmetries, is
the agent’s effort cannot be determined ex ante particularly relevant for USOs. It is generally
(Balakrishnan and Koza, 1993). This can lead to accepted that as a venture moves to later stages
adverse selection, or a “lemons” problem of development, the accompanied risk and
(Akerlof, 1970). Adverse selection refers to a uncertainty associated with that firm reduces
situation in which superior information is used to (Seppä and Laamanen, 2001). Shane (2004)
conduct immoral behavior prior to a transaction, defines early-stage USOs as being at a “minus
and Shane and Cable (2002) points out that this two” development stage, naturally leading to
often results in difficulties for entrepreneurs to extraordinary high uncertainty, and Wright et al.
obtain financing from investors because they (2006) state that informational asymmetries may
possess information about themselves and their be particularly significant in new high-tech
opportunities that potential investors do not firms. This has important implications for the
(Chan et al., 1990; Gompers, 1995). More suitability of traditional valuation methods – a
specifically, entrepreneurs are reluctant to discussion visited later in this study.
disclose all information regarding their
opportunity because it will make it easier for
other people to pursue the same opportunity,
Resource-Based Theory
which results in the investor receiving less In the 1959 resource-based view seminal text
information. Additionally, because of the “The theory of the growth of the firm”, Edith T.
superior information, entrepreneurs can engage Penrose introduces the notion of the firm as a
in opportunistic behavior (Shane and Cable, bundle of resources, and establishes causal links
2002). Consequently, Venkataraman (1997) between unique firm resources and capabilities,
writes that investors require the entrepreneur to and superior firm-level performance compared to
make large, irreversible and credible rivals (Kor and Mahoney, 2004). Wernerfelt
commitments to the venture, which drives up the (1984) writes that from a company's point of
entrepreneurs sunk cost, and drives high-quality view, resources and products are two sides of the
entrepreneurs from the market. This is what same coin, arguing that performance is directly
Akerlof (1970) denotes the “lemons” problem, affected by products, but indirectly (and
namely that only undesirable deals will be primarily) it is driven by the resources that go
available to investors. Figure 1 depicts the into producing them. He defined resources as
 Examining Equity Valuation of University Spin-off Companies: A literature review 5

“anything that could be thought of as a strength dependent on individual experiences, educational

or weakness for a given firm...”. In the 1991 backgrounds, social ties, initial financial capital,
article, Jay Barney expanded the definition of or cognitive ability to combine concepts and
resources to include “all assets, capabilities, information into new ideas. Further on, they
organizational processes, firm attributes, write that initial resource decisions made by the
information, knowledge etc. controlled by the entrepreneur has significant implications for
firm that enable it to conceive of and implement survival and growth. Brush et al. (2001) go on to
strategies that improve efficiency and define six categories of resources in new
effectiveness”. ventures: human, social, financial, physical,
technology, and organizational. Among these,
Building on the general assumption in resource-
human, technological, social and financial
based theory that resources and capabilities are
resources are highlighted as important for
heterogeneous across firms, Barney (1991)
research-based spinoffs (Mustar et al., 2006).
argues that sustained competitive advantages are
derived from the resources a company controls Human resources refer to the people in the
that are valuable, rare, imperfectly imitable company: attributes and experience of the
(VRIN) and non-substitutable. Valuable in the founding team, the management team and
sense that they enable the firm to implement personnel (Brush et al., 2001). According to the
strategies that improve its efficiency and upper echelon argument, new venture
effectiveness (Barney, 1991). Rare implying that performance, growth and success can be tied to
the resources must be scarce relative to demand. the individual attributes of the companies
Firms that possess resources that are both managerial group (Visintin and Pittino, 2014;
valuable and rare often have the first mover Eisenhardt and Schoonhoven, 1996).
advantage. However, to be a source of sustained
Social resources refer to the network of the
competitive advantage, they must also be
company, i.e. partners, industry and financial
imperfectly imitable, i.e. that other firms cannot
contacts of the founders (Brush et al., 2001).
copy them, and non-substitutable, meaning that
Eisenhardt and Schoonhoven (1996) write that
other resources cannot be used to implement
all action, including economic action, is
equivalent strategies (Barney, 1991).
embedded in a social fabric of opportunities, and
Underpinning this, Crook et al.
that interaction, and ultimately cooperation, is
(2008), conducting a meta-analysis on resources
likely to happen among people who know one
and performance, find that resources that meet
another. Such alliances can prove to be critical
the VRIN criteria are more strongly related to
resources (Hamel et al., 1989; Pisano and Teece,
performance that those that do not. Scholars have
1989) when firms are undertaking expensive or
however highlighted that the link between
risky strategies, as well as provide legitimacy
resources and performance is not only dependent
and market power (Wiewel and Hunter, 1985;
on the former being VRIN, but also on the
Hagedoorn, 1993; Baum and Oliver, 1991) that
organization's ability to capture the economic
improve strategic position. Stuart et al. (1999)
value that they can create (Barney and Clark,
write that third parties, such as investors and
2007). Mahoney and Pandian (1992) establish
employees, rely on the relationships of young
the missing link between resources and their
companies to assess their quality, and that young
utilization, and write that it is not merely the
companies “supported” by prominent
resources themselves that lead to competitive
relationships are more likely to perform well.
advantages, but rather the company’s distinct
This is because resource holders trust the ability
competencies in using the resources better than
of well-regarded organizations or individuals to
competitors.
assess quality under ambiguous circumstances
Building on the same principle as Mahoney and (Stuart, 1998), and that this “signaling effect”
Pandian (1992), Alvarez and Busenitz (2001) subsequently mitigates the risks of alliance
argue that it is through entrepreneurship and the formation such as poor implementation,
entrepreneurial process of discovering, opportunism, and appropriation of knowledge
understanding market opportunities, and (Larson, 1992). Eisenhardt and Schoonhoven
coordinating knowledge that inputs become (1996) go on to write that the advantages of top
heterogeneous outputs. Brush et al. (2001) write managers social resources are particularly
that entrepreneurs must first assemble resources, critical for young firms, because they have not
and then combine and develop them into had the time to establish firm-level networks.
resources that will yield distinctive capabilities.
Financial resources refer of the amount and type
Their ability to accomplish this will be
of funding a spinoff has received (Brush et al.,
6 Arif M. Mirza, Ole Rønning

Figure 2. Theoretical framework

2001). Brüderl et al. (1992) write that financial reviews (Cooper, 1988) are presented in the next
resources can act as buffers against random chapter. In this chapter the authors outline the
shocks, and Cooper et al. (1994) that capital search process.
allows entrepreneurs to learn and overcome
problems. Financial capital additionally enables Semi-Structured Literature Review
firms to pursue a broader range of activities as
well as more ambitious projects (Westhead et al., The semi-structured search is comprised of two
2001). parts: familiarizing with topics surrounding
USOs, and a review of general literature on new
Finally, technological resources refer to the venture valuation. For the former, the authors
company’s products and technology (Borch et started off by reading literature handed out by
al., 1999), as well as technical knowledge entrepreneurial scholars at NTNU,
associated with the two (Burgers et al., 2008), namely literature reviews (Mustar et al., 2006;
and intellectual property. Malerba and Orsenigo Rothaermel et al., 2007; Djokovic and Souitaris,
(1993) write that the factors related to the 2008; O’Shea et al., 2008; Rasmussen et al.,
endowed technological knowledge have an 2013; Wright, 2014), article (Widding et al.,
impact on the success of companies, and Shane 2009), and the book (Shane, 2004). For the latter,
and Cable (2002) that the strength of a firms the authors initially attempted to identify
technological endowments at founding is an literature on valuation of USOs through a
important predictor of its subsequent structured search, however the search yielded
performance. very limited results. Consequently, the focus was
With a basis in the above presented theory, the shifted to valuation in general, and a semi-
overarching argument that is employed in this structured search was chosen due to the vast and
study is that resource-based theory can be dispersed amount of literature available on the
leveraged to identify unique characteristics to subject. As a starting point, Fernandez (2002),
USOs. Further on, the distinction between risk recommended by scholars at NTNU, was read.
and uncertainty is used to analyze how these This book was used to identify relevant search
unique resource characteristics affect the terms, which are listed in table 1. A starting pool
perceived risk and uncertainty in a USO, and of articles were identified by searching on these
followingly how this affects the applicability of terms in the ISI Web of Science-database. To
traditional valuation methods. Figure 2 limit the scope of the search, only valuation
summarizes the theoretical framework. literature that fulfilled one of the following two
criteria was included in the pool of articles: 1)
the literature addressed the three most common
Methodology valuation methods used on new ventures, as
The research methodology employed in this identified by Fernandez (2002), or 2) the
study consists of a semi-structured literature literature addressed new venture valuation.
search on valuation methods, as well as a Thereafter snowballing was used (mainly
structured search on academic entrepreneurship backwards snowballing).
and USOs. The main results from the integrative
 Examining Equity Valuation of University Spin-off Companies: A literature review 7

The 60 different combinations of the terms in

Terms
table 2 were combined to create a search phrase1.
The search was limited to title, abstract and
Start%up
keywords, and returned 522 results. It would
have been possible to further limit the search
Firm
phrase by including terms describing the scope
of this thesis, i.e. “resources” or
Theory
“characteristics”, but to ensure that no articles
were missed due to the use of the “wrong” word,
Methods the authors chose to manually process the
articles with regards to scope. A set of discard
Venture5capitalist criteria was defined to ease the job of deciding
what articles to include in the literature review.
Entrepreneurial These criteria were as follows, and were used
throughout the article selection process:
Table 1. The search phrases used were a combination of
the terms listed in the table and the term “valuation*”. 1. Wrong unit of analysis 

Many of the retrieved articles focused on
Structured Literature Review a different level of analysis, for example
on the university itself, on government
Our structured review comprised of six steps:
policies or on TTOs. These articles were
Step 1 - Search. The first step of the review discarded.
comprised of a search in the ISI Web of Science-
2. Focus on aspects outside the scope of
database. The ISI database covers the leading this thesis
journals from a broad range of publishers, so the
A lot of the retrieved articles did not focus
chances of leaving out any important journals on the primary topic in this study, i.e.
were negligible. As previously highlighted, there
resources and resource characteristics.
are many names and terms used to describe the Examples of frequent topics were how
phenomenon of university spinoffs, and the
universities affect the creation of startups
initial semi-structured review was used to and intellectual property rights in spinoffs.
identify commonly used terms from among
These were discarded. 

others, Mustar et al. (2006), Rothaermel et al.
(2007), Djokovic and Souitaris (2008), O’Shea et 3. Not adhering to this study's
al. (2008), Rasmussen et al. (2013) and Wright definition of USOs
(2014). The different terms are listed in table 2. Articles that focused on firms that did not
qualify under the definition of USOs
Origin Nature+of+Firm employed in this study were
Academic New*venture*
discarded. 
This included articles that
covered other types of spinoffs, such as
Faculty Spin6off* those with corporations as the originating
institution.
Research6based Spin6out*
Step 2 - Title screening. The first screening of
Science6based Spinoff* the retrieved articles was based solely on reading
their title, and both authors reviewed the search
R&D*Management Spinout*
results to reduce subjective bias. Each author
Scientist New*firm* read the 522 titles and tagged articles believed to
fulfill the above criteria. To merge the authors
University Entrepreneurial individual lists the union of the two was taken.
This review left 190 articles.
Entrepreneurship
Step 3 - Abstract screening. The next stage
Start*up*
included a screening of the articles based on their
Spinning*out abstract. All abstracts were thoroughly reviewed,
and irrelevant articles in accordance with the
Table 2. USO search terms above screening criteria were discarded. After

1
Appendix 1
8 Arif M. Mirza, Ole Rønning

this, a total of 83 articles remained as relevant Descriptives

for this study.
The search revealed a total of 44 articles
Step 4 - Article retrieval. All the 83 identified published between 1986 and 2014, and scattered
articles were attempted collected, however, 23 across a total of 20 journals as shown in table 3,
articles turned out to be difficult to retrieve where Research Policy has the highest number of
despite their listing in the search results from the published articles, a total of 10. This observation
ISI database. Both authors conducted rigorous corresponds to the journal distribution found by
Google searches for the articles, and 7 of the Rothaermel et al. (2007), and gives this sample
missing articles were found. The remaining 16 validity. Table 3 also includes each journals
that were not found were primarily conference Impact Factor from 2013, which is an
proceedings. This left a total of 67 articles for unweighted factor that reflects the average
further assessment. number of citations of the most recent articles in
Step 5 - In-depth screening. All the 67 articles the journals (Garfield, 2006), and is mostly used
were subsequently given a quick read-through to to show the relative importance of a journal
determine if they were relevant for this study. within a field of research. Research Policy has a
The introductions, research questions and very high impact factor for a multi-disciplinary
conclusions were given particular focus. With a social science journal2, which further strengthens
basis in the three previously defined selection the validity of our results.
criteria, 26 articles were found to be relevant. #*of* Impact*
Journals:*20
articles Factor
Step 6 - Read-through. All the articles in the
Research(Policy 10 3,73
“relevant” category were read thoroughly.
The(Journal(of(Technology(Transfer 5 1,67
Backward snowballing was used to further
identify literature not uncovered by the literature Technovation 5 4,08

search. This stage involved reading the literature Journal(of(Business(Venturing 3 5,17

list in the articles already identified as relevant. R&D(Management 2 1,86
If a relevant citation was not included in the Technology(Analysis(&(Strategic(Management 2 1,05
literature search, the article was added to the list Entrepreneurship(Theory(and(Practice 2 3,92
of articles to read. The pool of literature on
Industrial(and(Corporate(Change 2 1,64
university spinoffs is limited in size, so
Technological(Forecasting(and(Social(Change 1 2,68
snowballing one or two steps was often enough
Journal(of(technology(management(&(
to return to already known literature. When all innovation
1 0,61
articles had been read, and all relevant citations
Economic(Development(Quarterly 1 0,81
were traced, the literature review ended up
Strategic(Entrepreneurship(Journal 1 2,48
containing 44 articles.
Journal(of(Management(Studies 1 4,35
Management(science 1 3,1
Economic(Inquiry 1 1,2
Science(and(Public(Policy 1 1,23
Tijdschrift(voor(economische(en(sociale(
1 1,27
geografie
The(business(of(biotechnology:(from(the(bench(
1 N/A
to(the(street
IEEE(Transactions(on(Engineering(Management 1 1,36
Venture(Capital 1 0,53

Table 3. Journal overview

Methodological Limitations
There are several limitations to the methodology
used in this study. As for the unstructured search,
it is possible that relevant articles on the topic of
valuation have been missed, which might have
been identified if a structured search was
employed. As for the structured search, only the
Figure 3. Literature search stages 2
http://www.journals.elsevier.com/research-policy
 Examining Equity Valuation of University Spin-off Companies: A literature review 9

ISI Web of Science database was used in our diversifiable, should warrant higher required
initial search. This excludes potential relevant returns. This is operationalized in the CAPM -
articles in journals not covered by the ISI model (capital asset pricing model), which
database, however the probability of this is low establishes that the required return on an
as the database covers the majority of the leading investment should be positively correlated to the
journals. Further on, backwards snowballing can long term risk-free interest rate, and to the stock
lead to the identification of articles in journals market premium (Wright Robbie, 1998):
not covered in the primary database used. A !" = !$ + &" (!( − !$ )+
second limitation to the structured search is the
limited set of keywords used to conduct the where
initial search. It is possible that certain relevant !$ = ,ℎ.+!/01+2!..+!34.+
terms might have been excluded, and
followingly relevant articles left unidentified. A &" = ,ℎ.+5.43+62+4ℎ.+0.78!/49+
third limitation relates to the method employed !( = :;<.74.=+>3!1.4+!.48!?+
to screen the initial 522 articles. The first
screening was based solely on titles, and even
though subjective bias was attempted reduced by In the context of valuation, the CAPM is
both authors reading all titles, it is possible that commonly used by investors to calculate their
false negatives can have occurred, i.e. relevant required return, and when valuing a company, it
articles eliminated. A fourth limitation is that is applied as the discount rate to estimate its
after the initial assessment of the titles, only one present value. Numerous studies have however
of the authors reviewed the abstracts for further highlighted that traditional risk/return models
elimination of articles. This bias could have been such as the CAPM are difficult to apply when
avoided if both authors reviewed the abstracts, valuing new ventures (Damodaran, 2009; Seppä
however, resource limitations prohibited this. A and Laamanen, 2001; Ge et al., 2005;
fifth limitation is that we narrowed our focus to Landström, 2007; Timmons and Spinelli, 2004).
literature exclusively focusing on USOs. By Whilst mature firms generally can exhibit several
excluding other research-based startups, which years of objective operating data, a new venture
have many similarities to USOs and are likely to cannot, which results in high uncertainty and
face many of the same valuation problems, we informational asymmetries (Sanders and Boivie,
might have overlooked relevant literature. 2004), and consequently increased expected
errors in forecasts (Wright Robbie, 1998).
Literature Review Compared to established firms, new ventures are
subject to several, qualitatively different sources
In this section, we present the main findings of risk and uncertainty (Berk et al., 2004), and
from the literature reviews on both valuation and particularly new technology based firms have
university spinoffs. First, a review of literature overall higher risks than other ventures (Brophy
on common valuation methods used by venture and Haessler, 1994). Whilst mature companies
capitalists as well as literature on new venture mainly face risks associated with product
valuation is introduced. Thereafter we present demand and production costs, new ventures also
literature on USO resources and resource face “technical” uncertainty regarding the
characteristics. success of the venture itself and the time and
cost required to bring the innovation to market,
Company Valuation as well as exogenous risks associated with the
actions of competitors and possible changes in
According to principles in corporate finance, the the market environment prior to launch (Berk et
return an investor should seek on an investment al., 2004). Despite these difficulties, early stage
is a function of the investments non-diversifiable company valuation remains an under researched
risk, measured through the investments Beta question, and studies that provide operational
(Brealey, 2012). The Beta is a measure of the guidance on valuing new ventures are lacking
investments correlation to the market, which (Ge et al., 2005). In fact, no relevant articles
represents the only risk an investor cannot aimed specifically at the valuation of early-stage
eliminate through diversifying her portfolio. high-tech firms or USOs were identified in our
Followingly, an investor should not require literature review. Admittedly, increasing
higher returns for holding investments that have attention within entrepreneurship literature has
unique or diversifiable risk, because it can be been paid to venture capital investments, but this
eliminated. Only market risk, which is non- has been on the criteria to screen a deal, i.e. the
10 Arif M. Mirza, Ole Rønning

Balance'Sheet Income'Statement Mixed'(Goodwill) Cash'Flow'Discounting Value'Creation Options

Book$value Multiples Classic Free$cash$flow EVA Black$and$Scholes

Union$of$European$
Adjusted$book$value PER Equity$cash$flow Economic$profit Investment$option
Accounting$Experts

Sales Dividens Cash$value$added Expand$the$project

Delay$the$
Liquidation$value P/EBITDA Abbreviated$income Capital$cash$flow
investment

Substantial$value Other$multiples Others APV CFROI Alternative$uses

Table 4. Valuation methods

criteria to invest or not, and not on the valuation future cash flows from the company discounted
process itself or how the criteria relate to firm at a risk adjusted rate (Petitt and Ferris, 2013).
value (Hudson, 2005; Ge et al., 2005).
“There appears to be wide
Due to the lacking literature on valuation of early agreement that, conceptually, the
stage high-tech firms and USOs, we focus our soundest measures of an assets
attention on common valuation methods used by value is the discounted value of the
investors on early stage ventures in general. future cash flow that it will
Further on we present associated weaknesses in generate “
the context of new venture valuation highlighted
– Lemke (1966)
in the reviewed literature. The lack of literature
on valuation of USOs confirms the lack of Cash flow discounting methods are based on
research attention this topic has received, and detailed forecasts of the company’s assets that
underpins the importance of this paper as the are related to the generation of cash flows, for
first of its kind. example interest payments, payroll, taxes,
revenues, materials etc. (Fernández, 2007).
Valuation Methods According to Damodaran (2009), there are four
key pieces that make up any valuation based on a
Fernandez (2002), in his book Company
discounted cash flow: existing assets and the
Valuation Methods and Shareholder Value
future cash flow generated by them; the expected
Creation, states that methods for valuing
growth rate of both existing assets and new
companies can be classified into six groups as
investments; the risk adjusted discount rate for
listed in table 4.
each of the cash flow streams; and an assessment
Because the focus of this study is on early stage of when the firm will become a stable growth
USOs, only the methods that are most commonly firm which is used to estimate the terminal value,
used by investors on early stage ventures are i.e. the value of all future cash flows after stable
presented, namely cash flow discounting, growth is reached. The general cash flow
balance sheet and income statement (Seppä and discounting method to estimate the present value
Laamanen, 2001). For each group of methods, of a firm can be summarized by the following
the authors give a brief introduction to the logic equation, where the last term represents the
underlying it, specific valuation techniques terminal value:
within each group, as well as limitations CDE CDG
emphasized by previous researchers on their @3A8.B = + +
1+! E !−H ∗ 1+! G
applicability to early stage ventures.
where
Cash Flow Discounting Valuation Methods CD/ = C30ℎ+2A6J+/?+9.3!+/
A well accepted axiom in mainstream finance is ! = !.K8/!.=+!.48!?+
that that the economic value of an investment is
the present value of all the future cash flows CD? = ,ℎ.+9.3!A9+730ℎ+2A6J+.;<.74.=
generated by it (Brealey, 2012). Discounted cash ++++++++++++++324.!+9.3!+?, J/4ℎ+3+H!6J4ℎ+!34.+H
flow methods rely on this axiom to define the
value of a company as the sum of the uncertain Estimating future cash flows can either be done
by a top-down approach, where one drills down
 Examining Equity Valuation of University Spin-off Companies: A literature review 11

from the total market for the company's service makes many strong assumptions that limit its
or product to estimate its expected revenues and usefulness.
earnings, or the bottom-up approach, where one
starts out considering the company's production Balance Sheet Methods
capacity. Discount rates are for mature public
Valuation methods based on balance sheets
firms often calculated using risk/return models
consider that the value of a company lies in its
such as the CAPM. In the case of new ventures,
assets, and therefore seek to value a company by
venture capitalists often apply predetermined
estimating asset values (Fernandez, 2002). The
discount rates, usually between 40% and 60%
most common balance sheet approach is the
(Sahlman, 1990). The high discount rates are
book value (or net asset) method, where the
justified by investors on the basis that they
company’s total liabilities are subtracted from its
include the risk free interest rate, the ventures
total assets to arrive at the company’s value. The
non-diversifiable risk, a market risk premium, an
adjusted book value method adjusts the values of
illiquidity discount, and finally compensation for
the company’s assets and liabilities according to
the value added by the investor (Sahlman, 1990).
their current market value. Company value
Further on, venture capitalists often use a
according to the liquidation method is the
modified version of the traditional cash flow
difference between value obtained from selling
discounting method known as the venture capital
the company’s assets and paying off its debt,
method. Instead of taking the firms perspective,
whilst the substantial value method seeks to
the venture capital method takes the perspective
identify the investment that must be made to
of the investor (Hellman, 2001). There are four
form a company having identical conditions and
basic steps involved; forecast sales or earnings
assets as the company being valued.
for a period of years, estimate the time at which
the investor will exit, value the company at the In contrast to more mature ventures who often
time of exit based on an assumed multiple, and have substantial physical assets, intellectual
finally apply an appropriate discount rate. assets such as patents may constitute a
Hellman (2001) points out that it is the simplicity significant part of the balance sheet of new
of the venture capital method that makes it technology based firm, and a number of studies
advantageous. have looked at the valuation of these. Parr and
Smith (2005) outline three quantitative methods
In the context of valuing new ventures, Keeley et
for valuing patents: the cost approach, the market
al. (1996) point out that high levels of risk,
approach and the income approach. The cost
which makes it difficult to estimate an
approach is based on the replacement cost of the
appropriate discount rate and to forecast cash
patented invention, the market approach uses
flows, as well as multiple investment stages,
comparable patent transactions as a basis to
which provides the investor options to abandon
obtain value, and lastly the income approach
the investment before making all planned
values a patent based on the estimated future
payments, makes it difficult to use the
income arising from it over its entire life (Parr
discounted cash flow methods. Damodaran
and Smith, 2005).
(2009) highlights that the absence of historical
data makes it difficult to estimate future Ge et al. (2005) write that the problem with
revenues, assess how revenues from existing balance sheet methods, when valuing early stage
assets will change under different companies, is that they ignore the value of
macroeconomic conditions as well as estimate growth opportunities. Additionally, most new
the expected return on invested capital. ventures have limited tangible assets, and
Fernández (2007) emphasizes that determining valuing intangible assets, such as patents for
the discount rate is one of the most important which there has not yet been identified a market
tasks in cash flow discounting methods, yet application, are according to Clarysse and
Damodaran (2009) writes that there is no way to Bruneel (2007) the most difficult issue in
estimate an equity beta, or use the market valuation. Fernandez (2002) adds that balance
interest rate on debt, rendering traditional risk sheet methods do not include factors such as
return models used to find discount rates useless. organizational issues or the current situation in
As for the venture capital method, Harper and the industry.
Rose (1993) characterize it as a “rule of
thumb”, Gompers and Lerner (2001) write that Income Statement Valuation Methods
the method is highly subjective and difficult to
These are methods that seek to value a company
justify, and Hellman (2001) concludes that it
based on sales, earnings or other indicators stated
12 Arif M. Mirza, Ole Rønning

on its income statement (Fernandez, 2002). or book value. Young companies tend not to
Some of the most common methods used are have scaling variables that are representative for
those that rely on different kinds of multiples, for how the company will perform when fully
example price/sales, where a company is valued operational. Fourth, equity claims vary across
based on its sales. Other common multiples are companies and affect the economic value of each
value of the company over EBIT, EBITDA, claim, and fifth, it is uncertain if geographical
operating cash flow or book value. Methods such differences render these methods erroneous. Ge
as these are also called relative valuation et al. (2005) finds three problems with using the
methods because the value is derived from what multiples approach on startups: lack of earnings,
the market is paying for similar firms difficulty in defining the boundaries for whom
(Damodaran, 2009), and the fundamental comparable companies are, and lastly, even with
argument underlying relative valuation is that a reference group defined, it is still quite
similar assets should be priced equally (Liu et subjective which multiple one should choose.
al., 2002). To use the multiples approach, three
“A biased analyst who is allowed to
parameters are required: a scaling factor, a
choose the multiple on which the
comparable multiple, and a discount rate. The
valuation is based and to pick the
scaling factor used varies greatly, and can be
comparable firms can essentially
sales, earnings, EBITDA or any other financial
ensure that almost any value can be
measure of the company being valued. The
justified.”
comparable multiples can either be based on
private or public transaction prices. Private – Damodaran (2001)
multiples are preferred by many valuation
analysts when valuing young ventures because Summary of New Venture Valuation
they claim that they are more representable in
Based on the above literature it is evident that
terms of illiquidity and risk. Illiquidity is a
new venture valuation is a difficult task, and that
central issue in new ventures due to few
the valuation methods most commonly used are
buyers/sellers and therefore substantial
in many cases inadequate for new ventures (Ge
transaction costs (Koeplin et al., 2000), and
et al., 2005; Fernandez, 2002; Damodaran,
should followingly be reflected in the new
2009). Yet, in most cases these methods are
ventures value. If a public multiple is used, the
readily applied, and in practice, many venture
analyst can choose an average multiple for the
capitalists simply use very high discount rates to
entire industry in question, or a multiple based
hedge against the uncertainty and inaccuracies of
on a single company. Finally, a discount rate
using these methods (Sahlman, 1990). This
may be necessary, if the scaling factor used is
however leads to fewer new ventures receiving
based on income statement forecasts. For new
funding, because venture capitalists hesitate to
ventures, it is often common to forecast an
fund any new venture that cannot, within
income statement that is representative of the
reasonable doubt, be expected to at least be
new ventures steady-state performance, i.e. when
valued between $25 to $50 million in five years
it has stable sales and earnings, and use a scaling
(MacMillan et al., 1986). When new venture
factor from this forecast. To calculate the present
valuation is already known to be problematic, to
value of the company it then becomes necessary
answer the research question in this study,
to discount the estimated value. If a public
namely the applicability of these methods on
multiple is used it is also common to add an
early-stage USOs, it then becomes the
illiquidity discount rate.
researcher's task to investigate how USOs differ
Damodaran (2009) writes that multiples are best from other new ventures, and how these
suited when valuing firms in businesses with differences affect the applicability of traditional
many other firms as well as where transactions valuation methods compared to other new
are common. Following, they are difficult to ventures. To identify these differences, the
apply to firms in unique businesses, and he authors review in the following chapter literature
highlights five potential problems with using on characteristics of initial resource endowments
multiples to value young businesses: first, of USOs, who together with the review of
comparative multiples stem from arms length literature on new venture valuation constitute the
transactions where side factors specific to the theoretical foundation for answering our research
transaction are hidden. Second, because private question.
transactions are infrequent, timing differences
occur. Third, to compare different companies, a
scaling variable is used, such as sales, earnings
 Examining Equity Valuation of University Spin-off Companies: A literature review 13

Resource Characteristics of Early Stage and technical knowledge is the core of all
University Spinoffs founded firms (Agarwal and Shah, 2014). Six
characteristics of initial technological resource
Mustar et al. (2006) state that four categories of endowments are highlighted in the reviewed
resources are especially important for research- literature; USOs tend to be founded on early
based spinoffs, namely human, technological, stage technologies that are radical and provide
social and financial. In the context of early-stage significant technical advances, have general-
valuation, the authors argue that technological purpose applications, that provide significant
resources are the main source of differences benefits to the customer, and finally that have
between a USO and other new ventures. This is strong intellectual property protection (Doutriaux
based on the fact that the literature review and Barker, 1995; Nelsen, 1991; del Campo et
revealed a majority of articles pertaining to the al., 1999; Shane, 2004). Followingly each
technological resources of a USO, and none or characteristic is reviewed.
few highlighting significant differences between
the other three resource categories compared to Early Stage Technologies
other new ventures. Also, as Agarwal and Shah
Licensing out unproven, early stage technology
(2014) point out, technology is the core of all
is difficult, and Doutriaux and Barker (1995) and
academic founded firms, and Shane (2004) that
Shane (2004) write that early stage inventions
these companies are often founded on years of
therefore tend to result in the creation of
research with nothing more than proof of
spinoffs. Further, these early stage technologies
concepts. The following review is therefore
have a number of implications on the new
heavily skewed towards a review of
venture. Studies show that the risk of loss
characteristics of initial technological resource
associated with investments from venture
endowments in USOs. It is worth mentioning
capitalists steadily decrease as the venture
that among the few findings on the other
evolves into later development stages (Seppä and
resource categories were Ensley and Hmieleski
Laamanen, 2001), which naturally leads to
(2005), who for human resources found that top
increased perceived risks associated with early
management teams (TMT) of USOs are more
stage technologies. Wright et al. (2006), citing
homogenous in terms of education, industry
Murray and Lott (1995) and Lockett et al.
experience, functional expertise and skills than
(2005), write that venture capitalists, especially
those of independent startups. Additionally, USO
in Europe, are reluctant to invest in early stage
founders tend to possess deeper technological
high-tech companies, and that informational
experience (Colombo and Piva, 2012), but little
asymmetries in companies working with
industry knowledge (Granovetter, 1973) and a
unproven technologies may be significant.
lower degree of entrepreneurial and managerial
Further, due to high levels of uncertainty
experience compared to those of nonacademic
regarding the technologies potential and the
new technology-based firms (Iacobucci et al.,
marketability of its functionality, common risk
2011; Colombo and Piva, 2012; Criaco et al.,
assessment measures are difficult to apply.
2013; Vohora et al., 2004; Agarwal and Shah,
2014). For social resources it was found that the “Many university inventions lead to
organizational environment of USOs the formation of spinoffs because
differentiate them from other independent they are early stage technologies
startups, where the parent university and the that are little more than ‘proofs of
associated institutes provide networks that are concept’ when the researcher
else hard to acquire. Through the parent discloses the invention to the
university, the spinoff can gain access to university technology licensing
laboratories and equipment as well as human office”
resources employed at the institution (Shane,
– Shane (2004)
2004). For financial resources it was found that
USOs face challenges in acquiring early stage
capital due to the uncertainty associated with
Radical Technologies That Provide Significant
these companies (Widding et al., 2009; Wright et Technical Advances
al., 2006; Shane, 2004) There are many definitions for what constitutes
radical innovation, and although the literature
Technological Resources has yet to converge to a universally accepted
Academic entrepreneurship is heavily based on definition, common to many of them is that the
technological advances in laboratory research, innovation incorporates technology that there is
14 Arif M. Mirza, Ole Rønning

little previous knowledge about and that is a spinoffs that possess general-purpose
clear and risky departure from existing practice technologies. This is because general purpose
(Green et al., 1995; Ettlie et al., 1984). Radical technology makes the new venture more suited
technologies are often recognized as engines of to adapt to changing circumstances, and
growth, productivity and performance subsequently provide a higher return to the
(Schumpeter, 1928; Linton and Walsh, 2008; investor. Clarysse et al. (2011) investigate how
Maine, 2008), and more often than not, technological knowledge-sources in USOs affect
university spinoffs are founded to exploit performance, and find that broad technology
technologies that are radical. Shane (2004) finds scopes are positively associated with growth, and
that for university spinoffs, it is important that therefore performance. They go on to write that
the underlying technology on which the this might be due to the fact that broader
company is founded represents transitions in the technologies allow the companies to shift
marketplace, or makes way for new product or between market applications when faced with
services, and Utterback (1996) states that dead ends. Finally, Increased performance due to
established firms with a dominant market broader scope of the technology may be linked to
position rarely adopt radical innovations due to increased possibilities for licensing out the
the fear that the new technology will cannibalize technology, either if the spinoff is not able to
their existing products or services. To leverage the technology themselves, or
commercialize university technology spinoffs simultaneously in different markets
become necessary, which leads to the (Gambardella and Giarratana, 2008).
technologies underlying these companies being
radical in nature. Further on, to offset the Significant Customer Value
uncertainty associated with radical technologies,
Because of the technology characteristics
the technologies need to represent a significant
mentioned above, commercialization of
technical advancement (del Campo et al., 1999)
university technology is a capital intensive task.
because they have greater economic value
This is further compounded by the inherent
(Harhoff et al., 1999). Spinoffs commercializing
challenges of bringing university technology to
radical technologies that do not provide
the market, such as distance to the end user
significant technical advances cannot be justified
because of an upstream position in the value
from an economic standpoint (Shane, 2004).
chain (Pavitt, 1984; Arora et al., 2001), and
“Several academic studies show requirements to other players in the value chain
that radical technologies tend to to adapt entirely new technology (Abernathy and
provide the basis for the creation of Clark, 1985; Tushman and Anderson, 1986).
university spinoffs, while This implies that creating a spinoff company
incremental technologies are more requires vast amount of financial resources, and
likely to be licensed by established university spinoffs cannot afford to exploit
companies” technologies that only offer small improvements
in customer value that will not provide sufficient
– Shane (2004)
financial returns. Shane (2004) followingly
writes that a characteristic of the initial
General-Purpose Technologies
technology endowments of USOs is that they
USOs tend to exploit technologies with broad provide significant customer value, because they
application areas and multiple fields of use are more likely to result in higher financial
(Nelsen, 1991; del Campo et al., 1999). This is returns (Harhoff et al., 1999). In addition,
primarily due to two reasons: first, applicability technologies that provide significant technical
in multiple markets render more market advances offset the increased uncertainty and
opportunities; and second, established companies risk associated with radical and unproven early
have trouble identifying what to use the stage technologies.
technologies for (Shane, 2004). General-purpose
technologies have important implications for the Strong Intellectual Property Protection
USO. Nelsen (1991) writes that general-purpose
Finally, strong intellectual property protection
technologies are positively correlated with
(IPR) is also a characteristic of the initial
spinoff performance because they allow spinoffs
technology resource endowments of USOs.
to pursue multiple application areas, hence
Nelsen (1991) states that strong intellectual
diversify risk, and multiple cash flow streams in
property (IP) protection is important for the
different points in the ventures development.
creation of USOs, as it is the only competitive
Shane (2004) writes that investors favor those
 Examining Equity Valuation of University Spin-off Companies: A literature review 15

advantage the firm has at inception, and Shane focus our efforts on exactly these characteristics
(2004) and De Coster and Butler (2005) support that set these companies apart, and we will
this by writing that spinoffs are generally more therefore primarily focus our discussion on
likely to be founded when the underlying technical resources. To further justify our focus
invention is protected by large portfolios of on technical resources we briefly comment on
broad-scoped patents. Patent portfolios are how the other three resource categories affect
valuable as they assert the spinoff greater control uncertainty in a USO compared to other new
over the technology, and patents with broad ventures in the context of valuation.
scopes allow for easier blocking of competitors.
In the reviewed literature, founding team
homogeneity in terms of backgrounds and
Summary of Literature Review experiences, was identified as a characteristic of
Based on the above literature reviews on venture human capital resources in USOs that differ from
valuation and USOs, it is evident that several other new ventures. There is obviously,
studies have separately focused on valuing early remembering the distinction between risk and
stage ventures, and the unique characteristics of uncertainty (Knight, 1921), uncertainty
initial resource endowments in USOs. More connected to this fact, such as whether the
specifically it has been seen that valuing these founding team has enough experience and
ventures is problematic and that these firms knowledge within all the necessary disciplines to
differ especially in their technological resource successfully launch the new venture. This is
endowments compared to other new ventures. categorized as uncertainty, not risk, because
However, no previous research, to our there is no possible way to estimate the
knowledge, have linked the two together and probability of the founding team having the
examined how the unique technological resource necessary combination of skills and experiences.
characteristics affect the process of valuing these However, this uncertainty can be significantly
firms. Different types of new ventures naturally reduced by investors through funding terms,
are faced with different types of challenges in the such as staging investments, hiring external
valuation process due to unique firm managers, or actively participating in the
characteristics, and ignoring these differences management team themselves, and so do not in
and assuming that the same methods can be any significant way affect the applicability of
applied to all companies may lead to significant traditional valuation methods on these firms.
erroneous valuations. In the following the The authors argue that social resources primarily
authors seek to fill this literature gap. assist in reducing uncertainty, and do not directly
affect the applicability of existing valuation
Discussion methods on early-stage USOs compared to other
new ventures. A social resource such as a
In light of the literature gap identified above, in relationship between the USO and an investor
this section we discuss how the characteristics of can significantly reduce informational
initial resource endowments affect the asymmetries between the two, effectively
applicability of traditional valuation methods reducing uncertainty, and making it easier for the
with a basis in the presented theoretical investor to value the company. However, the
framework. Because the underlying problem in lack of such a relationship does not increase
valuing a new venture is uncertainty, we uncertainty for an investor, as the outset is that
leverage the previously presented theory on risk no relationship exists. As such, there is no
and uncertainty (Knight, 1921) to first, argue significant difference between the social
how specific resource characteristics affect the resources in a USO and those in an independent
uncertainty associated with a USO, and second, startup that increase uncertainty in the context of
how that affects the applicability of traditional valuation.
valuation methods.
Finally, financial resources, or more precisely
University Spinoffs and Uncertainty the lack thereof, can affect the valuation of a
firm, but do not in any significant way contribute
The literature review revealed that particularly to increased uncertainty compared to other new
characteristics of the technical resources in these ventures. In the case of USOs it is likely that
companies are unique compared to other new several rounds of financing will be needed
ventures. When assessing whether valuation of before the new venture can survive by itself. The
early-stage USOs is indeed a more difficult task possible dilution effects of future financing
than valuing any other startup, it is necessary to rounds affect the valuations investors put on
16 Arif M. Mirza, Ole Rønning

companies, however, these effects can be statistical analysis, is an impossible task, leading
mitigated through contractual agreements to increased fundamental uncertainty. Finally,
(Sahlman, 1990), and should therefore not affect because of the two aforementioned reasons, the
the applicability of traditional valuation methods investor is faced with uncertainty in regards to
on these firms. the financial needs of the company. The longer
the development time frame, the larger the error
Technological Resources variance in the estimated financial needs is likely
to be. Big changes in financial needs after an
Six characteristics of USOs initial technological
initial investment can pressure the investor into a
resource endowments have been highlighted in
situation where they must keep investing to keep
the literature review, namely that the technology
their option on a future return, or they must face
is often early-stage, radical and provides
dilution. The probability of this happening is
significant technical advances, has general
likely to be smaller in companies that have
purpose applications, provides significant
shorter development timeframes, and
customer benefits, and has strong intellectual
followingly the early stages of the technology
property protection. In the following, we discuss
lead to USOs having higher uncertainty.
how each of these characteristics affect the
uncertainty in a USO from a valuation
Radical Technologies That Provide Significant
standpoint.
Technical Advances
Early Stage Technologies Technologies that are radical and provide
significant technical advances are by definition
USOs are often founded on research, and the
more risky to work with than mature
underlying technologies are rarely more than
technologies because they are previously
proof of concepts (Doutriaux and Barker, 1995).
untested and require new knowledge. USOs, who
It has been shown in the reviewed literature that
more often than not work with technologies that
early stage technologies are likely to have a
exhibit these characteristics, are venturing into
negative impact on performance due to higher
uncertain areas where knowledge is undeveloped
uncertainty and lower probability of receiving
or lacking, and the undertaking is costly (Green
funding, where the latter is previously shown to
et al., 1995). Further on, research has shown that
be the single most important determinant for
projects developing radical innovations are less
achieving IPO (Hayter, 2013; Clarysse et al.,
likely to be clearly tied to a market need, and
2011). The early stages of the technologies have
more likely to fail (Baker et al., 1985; Souder,
three important implications on the uncertainty
1987). Distinguishing between risk and
in USOs; it leads to uncertainty about the
uncertainty (Knight, 1921), radical projects that
commercial applicability of the research,
provide significant technical advances clearly
uncertainty regarding the development time
increase the uncertainty in a USO. It becomes
frame, and uncertainty about the financial needs
meaningless to try to estimate the probabilities of
of the company. First, because the underlying
a USO succeeding in developing the technology
factor for developing the technology in question
to commercial application, or the probability of
has been research at the university where the
finding a target market either a priori or through
USO originated, investors are faced with
empirical evaluation because the instances in
uncertainty as to whether the technology is
question are highly unique. The uncertainty in
applicable in a commercial product. Trying to
the USOs is further compounded by the fact that
estimate the probability of achieving this is an
these technologies represent new advanced
impossible task, and can therefore be categorized
knowledge areas, which investors have difficulty
as fundamental uncertainty. Second, turning the
understanding (Shane, 2004). This increases the
research into a commercial product is a time
informational asymmetries between the investor
consuming task, and research indicates that it
and the entrepreneur, making it even harder for
takes an average of 10 years to commercialize
the investor to assess the true potential of the
technology in a USO (Rasmussen et al., 2013;
project, and to assess the added value the
Lawton Smith and Ho, 2006). Significant
investor can deliver through other resources than
changes in the market environment and customer
capital, such as advice and guidance. The added
needs can occur in such a timeframe, which
uncertainty in USOs due to the technologies
increases the uncertainty as to whether the
these companies are founded on often being
investor will ever see a return on his investment.
radical and leading to significant advances, is
Again, estimating the probabilities of such
likely to decline as the venture matures to later
events occurring, either a priori or through
 Examining Equity Valuation of University Spin-off Companies: A literature review 17

stages of the development because more performance (Deeds et al., 2000; Zahra and
knowledge and information is accumulated, Bogner, 2000; Powers and McDougall, 2005).
however, in the early stages, the uncertainty seen From an investor’s viewpoint then, because they
from a valuation standpoint will, as we will are more likely to earn higher returns when the
discuss later, significantly impact investors USO has stronger intellectual property
ability to assess the project's value. protection, this characteristic reduces the
uncertainty in a USO. Further on, strong
General-Purpose Technologies intellectual property protection should offset
some of the uncertainty associated with long
General-purpose technologies are in previous
development times. By having strong IPR the
research highlighted as being associated with
USO can protect its technical space from
venture growth (Grant, 1996). The authors
competitors, and reduce the chances of other
further argue that general purpose technologies
companies beating them in commercializing the
in fact reduce uncertainty seen from the
same technology. Finally, if the USO is
viewpoint of an investor. USOs often work with
unsuccessful in commercializing the technology,
early-stage and radical technologies that do not
they will still assert control over it through the
have any target markets yet, and where there is
IPR, which can be sold off to other companies
uncertainty about ever finding an applicable
who might see value in owning the technology
market. This uncertainty is however reduced
for future use.
when the technology has broad application areas
and can be utilized in many different markets In the above discussion the authors have
and industries. This is substantiated by Nelsen presented how characteristics of initial technical
(1991), who points out that general-purpose resource endowments in USOs affect the
technologies are positively correlated with uncertainty in these companies seen from the
spinoff performance because they allow spinoffs viewpoint of an investor. The authors have
to diversify risk and earn multiple cash flows at argued that three of these characteristics, radical,
different points in time during the ventures significant technical advances and early stage, in
development. For example, general-purpose accordance with Knight's (1921) distinction
technologies permit the USO to license out the between risk and uncertainty, lead to high
technology to certain application areas whilst fundamental uncertainty, whilst on the other
keeping control over other application areas, hand, general purpose technologies that provide
which from an investor's viewpoint reduces the significant customer benefits and have strong
uncertainty about ever earning a return due to intellectual property protection in fact reduce it.
several parallel development efforts. In the following we discuss how this uncertainty
affects the applicability of traditional valuation
Significant Customer Value methods.
As for technologies that provide significant
customer benefits, no explanation is needed to University Spinoffs and Traditional
conclude that they reduce the uncertainty an Valuation Methods
investor is faced with when assessing a USO. The problems with using existing valuation
Technologies that are more worth to customers methods on startups in general have already been
are more likely to earn higher returns, and they highlighted in entrepreneurship and valuation
are not faced with the same “market push” literature (Ge et al., 2005; Fernandez, 2002;
problems that new technologies without a Damodaran, 2009). Despite this, these methods
specific demand face. It is therefore evident that are still commonly used amongst investors due
technologies that provide significant customer to the lack of better ways. Naturally, an estimate
benefits reduce the uncertainty associated with a is better than a guess. However, in the case of
USO. early stage USOs, using existing valuation
methods pose a new set of challenges due to the
Strong Intellectual Property Protection high uncertainty in these firms. Uncertainty is
Strong intellectual property protection is yet commonly factored into valuations by venture
another technical resource characteristic that, capitalists through high discount rates,
seen from the viewpoint of an investor, should commonly between 40% and 60% (Sahlman,
reduce uncertainty. Previous research on 1990). Using the same underlying logic, and
patenting has shown that there is a positive considering that a USO is highly more uncertain
correlation between patents and firm than the general startup, discount rates as high as
70%, 80% or 90% could be argued by an
18 Arif M. Mirza, Ole Rønning

investor. This however is in practice unfeasible, bottom-up approach, the starting point is the
and would result in almost all early-stage USOs focal company's capacity constraints, but
being categorized as unprofitable ventures. A because the company does not have defined
simple example can explain why. Assuming that products, it becomes meaningless to use this
a USO takes on average 10 years to method. Again, it is possible to assume what the
commercialize their technology, an investor, on capacity constraints can turn out to be if
average, has to wait 10 years before a return on successful in developing a product, but how do
their investment can be expected. Further on, you find the probability of succeeding?
assuming an investment of 1 MUSD for a 20%
Estimating the discount rate becomes an equally
equity stake, and that the investor requires a 35%
difficult task. Because of the technology’s early
yearly return over the investments lifetime
stage, investors cannot be certain if it is possible
(Zider, 1998), the present value of his investment
to utilize the technology in a product, or produce
needs to be 3,5M. Assuming a 70% discount
the technology at a commercial scale (Shane,
rate, the USO has to be valued at 3.5 billion USD
2004). Further on, this uncertainty is
after 10 years. Such a valuation is highly
compounded with the fact that the technology is
unlikely. Lets further assume that the investor
radical and represents significant technological
estimates to be able to sell his equity stake after
breakthroughs, which increases the probability of
5 years, the company still needs be valued at
failing in developing the technology into a
1.76 billion USD. These are very unlikely
commercial product or service. Obviously, this
valuations, and followingly, few or no USOs are
uncertainty should be included in a discount rate,
likely to attain funding if such a valuation
but how do you put a value on it? The situation
process is used. This example puts in perspective
in question is so highly unique, that it is simply
why these companies face financing problems
not possible to a priori or statistically estimate
(Shane, 2004). In the following we discuss how
the probability of finding an application area or
uncertainty affects the applicability of the three
succeeding in developing the technology to a
most common valuation methods previously
commercial scale, and it is therefore highly
presented.
difficult to account for in a risk adjusted rate.
Moreover, the example in the introduction
Cash Flow Discounting Methods
showed that the common way of handling
There are four key pieces required to use the uncertainty in new ventures by investors, namely
cash flow discounting method: existing assets applying high discount rates, is not possible in
and the future cash flow generated by them; the early-stage USOs. Finally, because of the early
expected growth rate of both existing assets and stages of the technology and the subsequent long
new investments; the risk adjusted discount rate development times these companies face, the
for each of the cash flow streams; and an time over which cash flows should be discounted
assessment of when the firm will become a as well as determining the terminal value is
stable growth firm which is used to estimate the highly challenging. It is likely, as Knight (1921)
terminal value. However, estimating any of these writes, that people, such as founders and
four key pieces is a significant challenge in early investors, will try to estimate their opinions in
stage USOs. terms of probability judgments, however,
because the instances in question are so highly
The first two criteria relate to estimating future
unique, and they therefore are facing unique
cash flows. For estimating future cash flows, the
challenges not previously solved, there is simply
literature review highlighted two common
no way to quantify this uncertainty. Clearly,
methods: the top-down and bottom-up approach.
estimating the parameters necessary to value an
With the top-down approach, one starts out with
early stage USO using discounted cash flow
the total market for the product, and then
valuation methods pose significant challenges
narrows down to the attainable market share of
due to high fundamental uncertainty, and these
the focal firm. This however, because the
models are therefore poorly suited for valuing
technology is radical and early stage and
such companies. This uncertainty is a result of
followingly does not have a defined market or
the initial technological resource endowments
application area (Wright et al., 2006; Vohora et
being early stage, radical and representing
al., 2004), is not possible. Indeed it can be
significant technological breakthroughs.
argued that the spinoff likely has mapped out
potential markets, but how do you weight the
sizes of each market to reach the total
addressable market for the firm? With the
 Examining Equity Valuation of University Spin-off Companies: A literature review 19

Balance Sheet Methods cutting edge technologies that are protected by

patents, finding a comparable firm with the same
Balance sheet valuation methods use the firm’s
technology characteristics is likely to prove
assets and liabilities to calculate its value. In the
difficult, rendering this group of valuation
case of early stage USOs, many are likely to not
methods unfeasible on early stage USOs.
have any valuable assets in place, and if they do
have assets, the majority of these are likely to be As shown in the above discussion, the fact that
in the form of intellectual property. As initial technological resource endowments in
previously stated, common methods for valuing USOs tend to be early stage, radical and
intellectual property are the cost approach, represent technological breakthroughs, increases
market approach and income approach. The cost the uncertainty in these firms significantly, and
approach values intellectual property based on make it difficult to estimate future cash flows,
the investment required to replace its future discount rates and find comparable firms, which
product or service capability, often using renders the three most common valuation
historical requirements to get to the current methods unfeasible on early stage USOs. The
technological development stage as well as authors argue that the early stage of the
expenses incurred. Because the technologies in technology, which implies that very limited
USOs in many cases do not have a defined future development towards a finished product has been
service or product stemming from the intellectual conducted, is the largest contributor to this
property, and additionally the research being a uncertainty, and in fact, when succeeding in
result of several years of work at the university reaching later stages of the development, i.e.
where researchers have access to a vast amount proving the technologies feasibility in a market
of resources such as laboratories with costly application, radicalness and significant
equipment (Shane, 2004) or experts in numerous technological advances may reduce the
fields, it is difficult to estimate the cost of uncertainty investors perceive when assessing
replicating the research. In the case of the market the company because the new venture controls a
approach, there are two requisites, an active superior patented technology. It is followingly
public market and an exchange of comparable natural to believe, that as the new venture
goods, neither of which are likely to exist for the develops into later stages, the uncertainty will be
unique and radical technologies which these significantly reduced, and traditional valuation
companies are founded on. An even more methods will again become applicable. Table 5
compelling argument against valuing the summarizes how characteristics of the initial
intellectual property of university spinoffs with technological resource endowments affect the
the use of the cost or market approach is that suitability of the above mentioned valuation
they don’t include the future revenues the methods.
intellectual property has the possibility to create,
which is where the real value of the patent lies in
the case of a USO. Finally, the income method is
Conclusion and Implications
based on valuing intellectual property on its Valuation is a central part in deciding whether
potential for generating future income, however, one should invest in a company or not (Keeley et
this requires the estimation of future cash flows al., 1996), and it is often one of the more
and discount rates, which has already been challenging discussions between an entrepreneur
shown above to pose significant challenges. and the investor (Quindlen, 2000). This study
investigates the applicability of traditional
Income Statement Methods valuation methods on the phenomenon of
university spinoffs, which in the last three
Valuation methods based on the firm’s income
decades have grown considerably in numbers
statement are the last group of methods (Rothaermel et al., 2007), and become an
presented in this study, and rely on an income
important part of the economic landscape
statement that is representative of the company’s (Shane, 2004; Bonardo et al., 2011; Rasmussen
long term performance, as well as the
et al., 2013). More specifically, this study
identification of what the market is paying for
leverages a theoretical framework comprised of
comparable firms. Here again, because of the
risk and uncertainty and resource-based theory to
early stages of the technology and the following
pinpoint initial resource characteristics of USOs
long development times, estimating what the
that render common valuation methods from
income statement will look like so far into the
corporate finance less applicable on these firms.
future is a nearly impossible task. Further on,
To accomplish this, an extensive literature
because these companies are working with
review was conducted to identify the most
20 Arif M. Mirza, Ole Rønning

Technology*Characteristic Balance*Sheet*methods Income*Statement*methods Discounted*Cash*Flow*methods

Early&stage Few-assets-on-balance-sheet,-and-does- Difficult-to-find-comparable-firms-because-of- Uncertainty-about-target-market-and-application-

not-incorporate-future-revenue-potential undefined-market/product area,-and-followingly-future-cash-flows

Intellectual-property-valuation- Uncertainty-regarding--steady&state-performance,- Uncertainty-regarding-technical-feasibility,-and-

dependent-on-cash-flow-estimation,- and-thereby-what-“scaling-factor”-to-use followingly-difficult-to-estimate-discount-rate
which-is-uncertain
Uncertainty-regarding-development-time,-and- Uncertainty-regarding-development-time,-and-
followingly-the-time-over-which-value-derived-from- followingly-the-time-over-which-cash-flows-should-be-
multiple-should-be-discounted discounted.
Uncertainty-about-scaling-technology-to-commercial-
production-

Radical Difficult-to-find-comparable-transactions- Difficult-to-find-comparable-firms-because-of-unique- Increases-uncertainty-regarding-technical-feasibility,-

to-value-intellectual-property technology and-followingly-difficult-to-estimate-discount-rate

Increases-uncertainty-regarding-technical-feasibility,- Uncertainty-about-scaling-technology-to-commercial-
and-followingly-difficult-to-estimate-discount-rate production

Significant-technical-advances Difficult-to-find-comparable-transactions- Difficult-to-find-comparable-firms-because-of-unique- Increases-uncertainty-regarding-technical-feasibility,-

to-value-intellectual-property technology and-followingly-difficult-to-estimate-discount-rate

Increases-uncertainty-regarding-technical-feasibility,- Uncertainty-about-scaling-technology-to-commercial-
and-followingly-difficult-to-estimate-discount-rate production

Table 5. USO characteristics and their effects on traditional valuation methods

common valuation methods used on new and at least be aware of their shortcomings. The
ventures, and the unique resource characteristics common denominator for why traditional
of USOs that separate these firms from other valuation methods are prone to error while used
new ventures. on early-stage USOs is the high fundamental
uncertainty, which makes it impossible to
The findings suggest that indeed, as Brush et al.
reliably estimate financial outputs necessary to
(2001) defines, the early stage resources of these
use these methods. This has big implications for
companies can be divided into four categories:
USOs, who face the risk of not attracting
human capital, social, financial technological.
necessary funding. The authors followingly
Further, using the distinction between risk and
recommend future research to investigate other
uncertainty as defined by Knight (1921), it has
valuation approaches than those of corporate
been shown that characteristics relating to the
finance, for example the applicability of
initial technological resource endowments, i.e.
valuation methods based on inputs such as
the early stage, radicalness and significant
resources, founder characteristics or networks,
technical advances, introduce high levels of
that can be objectively measured despite the high
uncertainty which render valuation methods
uncertainty associated with these firms, rather
based on cash flow discounting, balance sheets
than financial outputs. In our literature search, a
and income statements unfeasible on early stage
large body of literature was found on the
USOs. Additionally, the authors have briefly
relationship between resource endowments in
argued that it is the early stage of the technology
USOs and performance. This literature can be
that introduces the most uncertainty, and as the
leveraged to investigate the relationships
new venture progresses into more mature
between resources and firm value, with the
development stages and proves the feasibility of
underlying argument being that resources that
the technology, the uncertainty is likely to reduce
lead to higher long term performance should lead
significantly.
to higher valuations. We leave this task to future
The findings in this study has a vital implication research.
for valuation analysts and investors: using
common valuation methods on early-stage USOs References
may lead to erroneous valuations that do not
reflect the true value of the firm, and which ABERNATHY, W. J. & CLARK, K. B. 1985.
subsequently can lead to bad investment Innovation: Mapping the winds of creative
decisions. More importantly, they are likely to destruction. Research policy, 14, 3-22.
lead to many foregone investment opportunities, AGARWAL, R. & SHAH, S. K. 2014. Knowledge
and one should show caution when using them, sources of entrepreneurship: Firm
 Examining Equity Valuation of University Spin-off Companies: A literature review 21

formation by academic, user and advantage: The entrepreneurial challenge

employee innovators. Research Policy. of constructing a resource base. The
AKERLOF, G. A. 1970. The market for" lemons": Academy of Management Executive, 15,
Quality uncertainty and the market 64-78.
mechanism. The quarterly journal of BRÜDERL, J., PREISENDÖRFER, P. &
economics, 488-500. ZIEGLER, R. 1992. Survival chances of
ALVAREZ, S. A. & BUSENITZ, L. W. 2001. The newly founded business organizations.
entrepreneurship of resource-based theory. American sociological review, 227-242.
Journal of management, 27, 755-775. BURGERS, J. H., VAN DEN BOSCH, F. A. &
ARORA, A., FOSFURI, A. & GAMBARDELLA, VOLBERDA, H. W. 2008. Why new
A. 2001. Markets for technology: The business development projects fail: coping
economics of innovation and corporate with the differences of technological
strategy, MIT press. versus market knowledge. Long Range
BAKER, N., GREEN, S. & BEAN, A. 1985. How Planning, 41, 55-73.
management can influence the generation CAMERER, C. & WEBER, M. 1992. Recent
of ideas. Research Management, 28, 35- developments in modeling preferences:
42. Uncertainty and ambiguity. Journal of
BALAKRISHNAN, S. & KOZA, M. P. 1993. Risk and Uncertainty, 5, 325-370.
Information asymmetry, adverse selection CHAN, Y.-S., SIEGEL, D. & THAKOR, A. V.
and joint-ventures: Theory and evidence. 1990. Learning, corporate control and
Journal of Economic Behavior & performance requirements in venture
Organization, 20, 99-117. capital contracts. International Economic
BARNEY, J. 1991. Firm resources and sustained Review, 365-381.
competitive advantage. Journal of CLARYSSE, B. & BRUNEEL, J. 2007. Nurturing
management, 17, 99-120. and growing innovative start ups: the
BARNEY, J. B. & CLARK, D. N. 2007. Resource- role of policy as integrator. R&d
based theory: Creating and sustaining Management, 37, 139-149.
competitive advantage, Oxford University CLARYSSE, B., WRIGHT, M., LOCKETT, A.,
Press Oxford. MUSTAR, P. & KNOCKAERT, M. 2007.
BAUM, J. A. & OLIVER, C. 1991. Institutional Academic spin-offs, formal technology
linkages and organizational mortality. transfer and capital raising. Industrial and
Administrative science quarterly, 187-218. Corporate Change, 16, 609-640.
BERK, J. B., GREEN, R. C. & NAIK, V. 2004. CLARYSSE, B., WRIGHT, M. & VAN DE
Valuation and Return Dynamics of New VELDE, E. 2011. Entrepreneurial Origin,
Ventures. The Review of Financial Technological Knowledge, and the
Studies, 17, 1-35. Growth of Spin Off Companies. Journal
BIGLIARDI, B., GALATI, F. & VERBANO, C. of Management Studies, 48, 1420-1442.
2013. Evaluating performance of COHEN, B. 2006. Sustainable valley
university spin-off companies: Lessons entrepreneurial ecosystems. Business
from italy. Journal of technology Strategy and the Environment, 15, 1-14.
management & innovation, 8, 178-188. COLOMBO, M., MUSTAR, P. & WRIGHT, M.
BONARDO, D., PALEARI, S. & VISMARA, S. 2010. Dynamics of science-based
2011. Valuing University Based Firms: entrepreneurship. The Journal of
The Effects of Academic Affiliation on Technology Transfer, 35, 1-15.
IPO Performance. Entrepreneurship COLOMBO, M. G. & PIVA, E. 2012. Firms’
Theory and Practice, 35, 755-776. genetic characteristics and competence-
BORCH, O. J., HUSE, M. & SENNESETH, K. enlarging strategies: A comparison
1999. Resource configuration, competitive between academic and non-academic
strategies, and corporate entrepreneurship: high-tech start-ups. Research Policy, 41,
An empirical examination of small firms. 79-92.
Entrepreneurship Theory and Practice, COOPER, A. C., GIMENO-GASCON, F. J. &
24, 49-70. WOO, C. Y. 1994. Initial human and
BREALEY, R. A. 2012. Principles of corporate financial capital as predictors of new
finance, Tata McGraw-Hill Education. venture performance. Journal of business
BROPHY, D. J. & HAESSLER, M. R. 1994. venturing, 9, 371-395.
Evaluating the Venture Capital Fund COOPER, H. M. 1988. The integrative research
Agreement. Managerial Finance, 20, 53- review: A systematic aproach, Sage
64. Publications.
BRUSH, C. G., GREENE, P. G. & HART, M. M. CRIACO, G., MINOLA, T., MIGLIORINI, P. &
2001. From initial idea to unique SERAROLS-TARRÉS, C. 2013. “To have
22 Arif M. Mirza, Ole Rønning

and have not”: founders’ human capital incremental innovation. Management

and university start-up survival. The science, 30, 682-695.
Journal of Technology Transfer, 1-27. FAMA, E. F. 1991. Efficient capital markets: II.
CROOK, T. R., KETCHEN, D. J., COMBS, J. G. The journal of finance, 46, 1575-1617.
& TODD, S. Y. 2008. Strategic resources FERNANDEZ, P. 2002. Valuation methods and
and performance: a meta-analysis. shareholder value creation, Academic
Strategic Management Journal, 29, 1141- Press.
1154. FERNÁNDEZ, P. 2007. Company valuation
DAMODARAN, A. 2001. Corporate finance: methods. The most common errors in
theory and practice. valuations. Research Paper no, 449.
DAMODARAN, A. 2009. Valuing young, start-up FESTEL, G. 2013. Academic spin-offs, corporate
and growth companies: estimation issues spin-outs and company internal start-ups
and valuation challenges. Stern School of as technology transfer approach. The
Business, New York University (New Journal of Technology Transfer, 38, 454-
York). 470.
DE COSTER, R. & BUTLER, C. 2005. FULLER, A. W. & ROTHAERMEL, F. T. 2012.
Assessment of proposals for new When stars shine: the effects of faculty
technology ventures in the UK: founders on new technology ventures.
characteristics of university spin-off Strategic Entrepreneurship Journal, 6,
companies. Technovation, 25, 535-543. 220-235.
DEEDS, D. L., DECAROLIS, D. & COOMBS, J. GAMBARDELLA, A. & GIARRATANA, M.
2000. Dynamic capabilities and new 2008. General technologies, product
product development in high technology market fragmentation, and markets for
ventures: an empirical analysis of new technology: evidence from the software
biotechnology firms. Journal of Business security industry. Unpublished
venturing, 15, 211-229. manuscript.
DEL CAMPO, A. A., SPARKS, A., HILL, R. C. & GARFIELD, E. 2006. The history and meaning of
KELLER, R. T. 1999. The transfer and the journal impact factor. Jama, 295, 90-
commercialization of university- 93.
developed medical imaging technology: GE, D., MAHONEY, J. M. & MAHONEY, J. T.
opportunities and problems. Engineering 2005. New Venture Valuation by Venture
Management, IEEE Transactions on, 46, Capitalists: An Integrative Approach.
289-298. University of Illinois at Urbana-
DEQUECH, D. 2000. Fundamental uncertainty and Champaign, College of Business.
ambiguity. Eastern Economic Journal, 41- GOMPERS, P. & LERNER, J. 2001. The venture
60. capital revolution. Journal of economic
DJOKOVIC, D. & SOUITARIS, V. 2008. Spinouts perspectives, 145-168.
from academic institutions: a literature GOMPERS, P. A. 1995. Optimal investment,
review with suggestions for further monitoring, and the staging of venture
research. The Journal of Technology capital. The journal of finance, 50, 1461-
Transfer, 33, 225-247. 1489.
DOUTRIAUX, J. & BARKER, M. 1995. The GRANOVETTER, M. S. 1973. The strength of
university-industry relationship in science weak ties. American journal of sociology,
and technology. Gouvernement du 1360-1380.
Canada-Industry Canada. GRANT, R. M. 1996. Prospering in dynamically-
EISENHARDT, K. M. & SCHOONHOVEN, C. B. competitive environments: Organizational
1996. Resource-based view of strategic capability as knowledge integration.
alliance formation: Strategic and social Organization science, 7, 375-387.
effects in entrepreneurial firms. GREEN, S. G., GAVIN, M. B. & AIMAN-SMITH,
organization Science, 7, 136-150. L. 1995. Assessing a multidimensional
ENSLEY, M. D. & HMIELESKI, K. M. 2005. A measure of radical technological
comparative study of new venture top innovation. Engineering Management,
management team composition, dynamics IEEE Transactions on, 42, 203-214.
and performance between university- HAGEDOORN, J. 1993. Understanding the
based and independent start-ups. Research rationale of strategic technology
Policy, 34, 1091-1105. partnering: Nterorganizational modes of
ETTLIE, J. E., BRIDGES, W. P. & O'KEEFE, R. cooperation and sectoral differences.
D. 1984. Organization strategy and Strategic management journal, 14, 371-
structural differences for radical versus 385.
 Examining Equity Valuation of University Spin-off Companies: A literature review 23

HAMEL, G., DOZ, Y. L. & PRAHALAD, C. K. LINTON, J. D. & WALSH, S. T. 2008.

1989. Collaborate with your competitors Acceleration and extension of opportunity
and win. Harvard business review, 67, recognition for nanotechnologies and
133-139. other emerging technologies. International
HARHOFF, D., NARIN, F., SCHERER, F. M. & Small Business Journal, 26, 83-99.
VOPEL, K. 1999. Citation frequency and LIU, J., NISSIM, D. & THOMAS, J. 2002. Equity
the value of patented inventions. Review Valuation Using Multiples. Journal of
of Economics and statistics, 81, 511-515. Accounting Research, 40, 135-172.
HARPER, C. P. & ROSE, L. C. 1993. Accuracy of LOCKETT, A., SIEGEL, D., WRIGHT, M. &
appraisers and appraisal methods of ENSLEY, M. D. 2005. The creation of
closely held companies. spin-off firms at public research
HAYTER, C. S. 2013. Harnessing University institutions: Managerial and policy
Entrepreneurship for Economic Growth implications. Research Policy, 34, 981-
Factors of Success Among University 993.
Spin-offs. Economic Development MACMILLAN, I. C., SIEGEL, R. &
Quarterly, 27, 18-28. NARASIMHA, P. S. 1986. Criteria used
HELLMAN, T. 2001. A note on Valuation of by venture capitalists to evaluate new
venture Capital Deals. Harvard Business venture proposals. Journal of Business
Review. venturing, 1, 119-128.
HUDSON, E. 2005. A Review of Research into MAHONEY, J. T. & PANDIAN, J. R. 1992. The
Venture Capitalists' Decision Making: resource based view within the
Implications for Entrepreneurs, Venture conversation of strategic management.
Capitalists and Researchers. Journal of Strategic management journal, 13, 363-
Economic and Social Policy, 10, 3. 380.
IACOBUCCI, D., IACOPINI, A., MICOZZI, A. & MAINE, E. 2008. Radical innovation through
ORSINI, S. 2011. Fostering internal corporate venturing: Degussa's
entrepreneurship in academic spin-offs. commercialization of nanomaterials. R&d
International Journal of Entrepreneurship Management, 38, 359-371.
and Small Business, 12, 513-533. MALERBA, F. & ORSENIGO, L. 1993.
KEELEY, R. H., PUNJABI, S. & TURKI, L. 1996. Technological regimes and firm bebavior.
Valuation of early-stage ventures: option Industrial and corporate change, 2, 45-71.
valuation models vs. traditional MARION, T. J., DUNLAP, D. R. & FRIAR, J. H.
approaches. The Journal of 2012. The university entrepreneur: a
Entrepreneurial Finance, 5, 115-138. census and survey of attributes and
KNIGHT, F. H. 1921. Risk, uncertainty and profit. outcomes. R&D Management, 42, 401-
New York: Hart, Schaffner and Marx. 419.
KOEPLIN, J., SARIN, A. & SHAPIRO, A. C. MILLER, D. 2003. An asymmetry based view of
2000. THE PRIVATE COMPANY advantage: towards an attainable
DISCOUNT. Journal of Applied sustainability. Strategic management
Corporate Finance, 12, 94-101. journal, 24, 961-976.
KOR, Y. Y. & MAHONEY, J. T. 2004. Edith MURRAY, G. C. & LOTT, J. 1995. Have UK
Penrose's (1959) Contributions to the venture capitalists a bias against
Resource-based View of Strategic investment in new technology-based
Management. Journal of Management firms? Research Policy, 24, 283-299.
Studies, 41, 183-191. MUSTAR, P., RENAULT, M., COLOMBO, M.
LANDSTRÖM, H. 2007. Handbook of research on G., PIVA, E., FONTES, M., LOCKETT,
venture capital, Edward Elgar Publishing. A., WRIGHT, M., CLARYSSE, B. &
LARSON, A. 1992. Network dyads in MORAY, N. 2006. Conceptualising the
entrepreneurial settings: A study of the heterogeneity of research-based spin-offs:
governance of exchange relationships. A multi-dimensional taxonomy. Research
Administrative science quarterly, 76-104. policy, 35, 289-308.
LAWTON SMITH, H. & HO, K. 2006. Measuring NECK, H. M., MEYER, G. D., COHEN, B. &
the performance of Oxford University, CORBETT, A. C. 2004. An
Oxford Brookes University and the entrepreneurial system view of new
government laboratories’ spin-off venture creation. Journal of Small
companies. Research Policy, 35, 1554- Business Management, 42, 190-208.
1568. NELSEN, L. L. 1991. The lifeblood of
LEMKE, K. W. 1966. Asset Valuation and Income biotechnology: University-industry
Theory. The Accounting Review, 41, 32- technology transfer. The business of
41.
24 Arif M. Mirza, Ole Rønning

biotechnology: from the bench to the evidence. R&D Management, 31, 215-
street, 39-75. 230.
O’SHEA, R. P., CHUGH, H. & ALLEN, T. J. SHANE, S. 2004. Academic entrepreneurship:
2008. Determinants and consequences of University spinoffs and wealth creation,
university spinoff activity: a conceptual Edward Elgar Publishing.
framework. The Journal of Technology SHANE, S. & CABLE, D. 2002. Network Ties,
Transfer, 33, 653-666. Reputation, and the Financing of New
PARR, R. L. & SMITH, G. V. 2005. Intellectual Ventures. Management Science, 48, 364-
Property: Valuation, Exploitation, and 381.
Infringement Damages, Wiley. SOUDER, W. E. 1987. Managing new product
PAVITT, K. 1984. Sectoral patterns of technical innovations, Lexington books Lexington,
change: towards a taxonomy and a theory. MA.
Research policy, 13, 343-373. STUART, T. E. 1998. Network positions and
PETITT, B. & FERRIS, K. R. 2013. Valuation for propensities to collaborate: An
mergers and acquisitions, Pearson investigation of strategic alliance
Education. formation in a high-technology industry.
PISANO, G. P. & TEECE, D. 1989. Collaborative Administrative Science Quarterly, 668-
Arrangements and Global Technology 698.
Strategy: Some Evidence from the STUART, T. E., HOANG, H. & HYBELS, R. C.
Telecommunications Equipment Industry. 1999. Interorganizational endorsements
Research on Technological Innovation, and the performance of entrepreneurial
Management, and Policy, 4. ventures. Administrative science
POWERS, J. B. & MCDOUGALL, P. P. 2005. quarterly, 44, 315-349.
University start-up formation and SÆTRE, A. S., ATKINSON, O. T. & ELLERÅS,
technology licensing with firms that go B. 2006. University spin-offs as
public: a resource-based view of academic technology commercialization: a
entrepreneurship. Journal of Business comparative study between Norway,
Venturing, 20, 291-311. Sweden and the UnitedStates. Trondheim:
PRESSMAN, L. 2002. AUTM Licensing Survey: NTNU, Department of Industrial
FY 2001. Northbrook, IL: Association of Economics and Technology Management.
University Technology Managers. TIMMONS, J. A. & SPINELLI, S. 2004. New
QUINDLEN, R. 2000. Confessions of a venture venture creation: entrepreneurship for the
capitalist: inside the high-stakes world of 21st century, McGraw-Hill/Irwin.
start-up financing, Hachette Digital, Inc. TUSHMAN, M. L. & ANDERSON, P. 1986.
RAPPERT, B., WEBSTER, A. & CHARLES, D. Technological discontinuities and
1999. Making sense of diversity and organizational environments.
reluctance: academic–industrial relations Administrative science quarterly, 439-465.
and intellectual property. Research policy, TYEBJEE, T. T. & BRUNO, A. V. 1984. A model
28, 873-890. of venture capitalist investment activity.
RASMUSSEN, E., BORLAUG, S. B. & Management science, 30, 1051-1066.
BULANOVA, O. 2013. Verdiskaping i UTTERBACK, J. M. 1996. Mastering the
forskningsbaserte selskaper og lisenser dynamics of innovation, Harvard Business
støttet av FORNY-programmet. Press.
ROTHAERMEL, F. T., AGUNG, S. D. & JIANG, VENKATARAMAN, S. 1997. The distinctive
L. 2007. University entrepreneurship: a domain of entrepreneurship research.
taxonomy of the literature. Industrial and Advances in entrepreneurship, firm
corporate change, 16, 691-791. emergence and growth, 3, 119-138.
SAHLMAN, W. A. 1990. The structure and VISINTIN, F. & PITTINO, D. 2014. Founding
governance of venture-capital team composition and early performance
organizations. Journal of financial of university—Based spin-off companies.
economics, 27, 473-521. Technovation, 34, 31-43.
SANDERS, W. G. & BOIVIE, S. 2004. Sorting VOHORA, A., LOCKETT, A. & WRIGHT, M.
things out: valuation of new firms in Critical junctures in the growth in
uncertain markets. Strategic Management university high-tech spinout companies.
Journal, 25, 167-186. Conference paper at Frontiers of
SCHUMPETER, J. 1928. The instability of Entrepreneurship Research, University of
capitalism. The economic journal, 361- Colorado, 2002.
386. VOHORA, A., WRIGHT, M. & LOCKETT, A.
SEPPÄ, T. J. & LAAMANEN, T. 2001. Valuation 2004. Critical junctures in the
of venture capital investments: empirical development of university high-tech
 Examining Equity Valuation of University Spin-off Companies: A literature review 25

spinout companies. Research policy, 33,

147-175.
WERNERFELT, B. 1984. A resource based view
of the firm. Strategic management
journal, 5, 171-180.
WESTHEAD, P., WRIGHT, M. & UCBASARAN,
D. 2001. The internationalization of new
and small firms: A resource-based view.
Journal of Business Venturing, 16, 333-
358.
WIDDING, L. O., MATHISEN, M. T. &
MADSEN, O. 2009. University-affiliated
venture capital funds: Funding of
university spin-off companies.
International Journal of Technology
Transfer and Commercialisation, 8, 229-
245.
WIEWEL, W. & HUNTER, A. 1985. The
interorganizational network as a resource:
A comparative case study on
organizational genesis. Administrative
Science Quarterly, 482-496.
WRIGHT, M. 2014. Academic entrepreneurship,
technology transfer and society: where
next? The Journal of Technology Transfer,
39, 322-334.
WRIGHT, M., CLARYSSE, B. & MOSEY, S.
2012. Strategic entrepreneurship, resource
orchestration and growing spin-offs from
universities. Technology Analysis &
Strategic Management, 24, 911-927.
WRIGHT, M., LOCKETT, A., CLARYSSE, B. &
BINKS, M. 2006. University spin-out
companies and venture capital. Research
Policy, 35, 481-501.
WRIGHT ROBBIE, M. K. 1998. Venture Capital
and Private Equity: A Review and
Synthesis. Journal of Business Finance &
Accounting, 25, 521-570.
ZAHRA, S. A. & BOGNER, W. C. 2000.
Technology strategy and software new
ventures' performance: exploring the
moderating effect of the competitive
environment. Journal of business
venturing, 15, 135-173.
ZIDER, B. 1998. How venture capital works.
Harvard business review, 76, 131-139.
!
 Resource-based Equity Valuation of University
Spin-off Companies: An Empirical Study
Arif M. Mirza, Ole Rønning

ABSTRACT
Company valuation is a critical aspect in any deal concerning funding to new ventures. However, valuing
an early-stage university spin-off (USO) using traditional valuation methods from corporate finance is an
inherently difficult task. These are companies spun out of universities, often with a basis in early-stage and
radical technologies which provide significant technical advances. These technological characteristics lead
to high fundamental uncertainty, which makes it difficult to reliably apply valuation methods such as cash
flow discounting or relative valuation. This is believed to be one of the reasons these companies struggle to
receive financing in their critical early stages. This study leverages research from strategic management in
a valuation perspective, and develops a conceptual valuation framework based on inputs (e.g. resources)
rather than financial outputs (e.g. cash flows, balance sheets etc.). To test the framework, six hypotheses
about the relationship between initial resource endowments and the long term equity value of an USO are
developed with a basis in existing research on resource-performance. A sample of 63 Norwegian USOs are
used to develop an empirical model to test the hypotheses. It is found that the degree of heterogeneity in
the founding teams educational backgrounds and previous work experiences, the quality of the parent
university, and the number of filed and published patents in the early stages of a USO are all positively
related to its long term equity value. It is followingly argued that USOs who, ceteris paribus, exhibit these
resource characteristics in their early stages should receive higher valuations than those that do not. Further
on, a valuation framework based on resources as inputs shows promise, however significant future
research is required to develop such a tool.

an early-stage university spin-off (USO) even

Introduction harder. Widding et al. (2009) state that high
levels of uncertainty and capital requirements in
Valuation is a central part in deciding whether
these companies result in investors not having
one should invest in a company or not (Keeley et
sufficient information to find suitable terms for
al., 1996). For venture capitalists, valuation is
funding. Mirza and Rønning (2015), examining
important because it determines the proportion of
the applicability of traditional valuation methods
shares they receive in return for their
on early-stage USOs, conclude that the early
investments, guides the overall profitability of
stage, the radicalness, and the significant
their fund and thus also affects their relationship
technical advancements offered by the
with their fund providers (Miloud et al., 2012).
technologies these companies are founded on
For entrepreneurs, the valuation is important
lead to high fundamental uncertainty, and
because it sets a value on the efforts and
followingly that valuing a USO using one of the
resources they have put into the new venture,
three most common valuation methods, i.e. cash
and the portion of the firm they must give up to
flow discounting, balance sheet and income
obtain external financing. It has also been shown
statement, is infeasible. The infeasibility of these
that fair valuation aligns the ambition of both the
methods can in large be attributed to the lack of
investor and entrepreneur, and that it helps to
operational history necessary to apply them. The
structure the deal between the two through the
outputs that these methods are dependent on,
identification of reasonable financing terms (De
such as cash flows and balance sheets, are non
Clercq et al., 2006).
existent for a USO in its early stages, and further
Valuing a young venture is however known to be on, because of high fundamental uncertainty,
an inherently difficult task (Timmons and they are impossible to reliably estimate.
Spinelli, 2004; Damodaran, 2009), and valuing Moreover, as Mirza and Rønning (2015) shows,
2 Arif M. Mirza, Ole Rønning

the common method of applying discount rates performance and equity valuation together. We
as high as 80% to hedge against this uncertainty review existing literature and develop an
results in almost all early-stage USOs being analytical model to investigate the following
categorized as unprofitable investments. These research question:
valuation issues are likely to be more prevalent
RQ: How can initial resource
in USOs founded on technology, which require
endowments be used to assess the
more capital, rather than knowledge (i.e.
early-stage value of a USO?
consultancy companies). Following Mirza and
Rønning (2015), we therefore define early stage The above research question can further be
USOs as companies that involve transfer of justified by two important reasons. First, USOs
intellectual property rights to the new venture, are ventures that require significant amounts of
and that are either developed by faculty members capital in their early stages (Shane, 2004;
based on their own research, or created Clarysse and Bruneel, 2007; Neck et al., 2004),
specifically to capitalize on academic research but that find it difficult to acquire the sufficient
(Shane, 2001; Colombo et al., 2010; Long, amount of funding. This is partially due to the
2002). As for their early stage nature we assume long development times associated with USOs
them to be in the opportunity framing phase in (Shane, 2004; Rasmussen et al., 2012), where
Vohora et al. (2002) five phase development investors prefer a short time to market, as well as
model, which is where USOs first face the issue investors not being able to adequately assess the
of raising capital. value of the companies due to uncertainty.
Through this study the authors therefore hope to
In the absence of the financial outputs necessary
establish the initial theoretical linkages for
to apply existing valuation methods, following in
developing a valuation framework with which
the footsteps of Ge et al. (2005), we propose to
investors can adequately assess the value of
leverage theories within strategic management
early-stage USOs, and thereby reduce the
that help predict and explain firm performance to
financing gap these companies experience
value a company in its early stages. While
(Shane, 2004).
traditional corporate finance has focused on the
use of cash flows and returns, fields within Second, USOs, when successful, generate
strategic management have focused on the significant economic value (Shane, 2004), and
conditions under which superior returns can be the authors believe that better valuation methods
achieved (Barney et al., 2001; Kor and Mahoney, may not only lead to increased flow of financing
2004; Scherer and Ross, 1990), and the latter as well as more appropriate funding terms for
should therefore in theory be suited to identify these firms, but also greater wealth creation in
variables important for valuation when the society. Bonardo et al. (2011), for example,
former cannot be measured. Indeed, when it is found that one-fourth of the high-tech SMEs that
difficult to value a company based on outputs went public in Europe during the period 1995-
(cash flows etc.), a method based on inputs 2003 were USOs, and concludes that university-
(resources, founder characteristics, network etc.) based firms represent a significant contribution
that can be objectively measured may prove to European financial markets.
more satisfactory (Ge et al., 2005).
This study is divided into five parts, where this
Within the research on USOs, links between introduction constitutes the first part. In the
initial resource endowments and subsequent second part we develop hypotheses with a basis
performance has received notable attention in resource-based theory and resource-
(Rasmussen et al., 2012). Based on this research performance relationships highlighted in existing
the authors investigate whether these initial research streams. In the third part we present the
resource endowments can be used as predictors analytical methodology employed to test our
in the early stages of a USO for what the long hypotheses, and finally in part four we present
term value of the firm will be, and as such lay and discuss our results and finish off with
the foundation for a resource-based valuation implications and areas for future research in part
framework. It is not the goal of the authors to five.
develop an entire valuation framework. Rather
we seek to investigate its feasibility, and develop Theoretical Framework and
a conceptual framework to lay the stepping stone
for future research on the topic. This study, Hypotheses
which to our knowledge is the first of its kind,
Mirza and Rønning (2015) find that high
aims to bring the research on university spin-off
fundamental uncertainty in USOs makes it
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 3

Figure 1. Conceptual framework

difficult to reliably forecast financial outputs, and the Industrial Organization perspective,
which renders traditional valuation methods which emphasizes external forces and the
difficult to apply. The authors argue that the structure of the market as the driver of
forecasting process can in its simplest form be performance (Porter, 1981). The authors argue
broken down to a problem of making an that the former of these is best suited in the
assessment under uncertainty, and from context of valuing early-stage USOs. Whilst the
cognitive psychology it is known that to cope latter is dependent on many factors outside the
with such uncertainty, heuristics are commonly new ventures control and can rapidly change,
applied (Gigerenzer et al., 2011). Heuristics are resources define and distinguish a company from
efficient cognitive processes, or mental others (Miller, 2003; Penrose, 1995), are
shortcuts, used to make decisions without objectively measurable despite uncertainty, and
assessing all the necessary information one help explain superior firm performance (Kor and
would want to make an optimal decision, and Mahoney, 2004). As Barney (1991) writes,
research on the use of heuristics has in fact resources that are valuable, rare, inimitable and
shown that they can lead to more accurate nonsubstitutable (VRIN) have the potential to
judgments than weighing and analyzing create sustained competitive advantages, if the
information in an uncertain situation (Gigerenzer company has the necessary capabilities to utilize
and Gaissmaier, 2011). Gigerenzer et al. (2011) them in a value-creating manner (Mahoney and
write that heuristics exploit the information Pandian, 1992; Alvarez and Busenitz, 2001;
structure of the environment, and that they in a Brush et al., 2001). Theoretically speaking then,
world full of uncertainties and surprises often a firm with certain resources, when utilized the
lead to more accurate assessments than complex right way, should, ceteris paribus, perform better
methods. than firms that do not have the same resources
because of their competitive advantages. These
In the context of valuation, instead of calculating
advantages should lead to higher financial
highly uncertain cash flows, or estimating
returns, and subsequently higher firm values. A
probabilities that are impossible to estimate, it
firm that exhibits these resources in its early
then seems logical to cope with uncertainty in
stages should followingly be valued higher than
the same manner, and value early-stage USOs by
a firm that does not. The obvious limitation to
applying heuristics. To identify such heuristics,
this approach is that the performance impact of
it becomes necessary to look towards theories
resources are dependent on the USOs capability
from strategic management that help predict firm
to utilize them (Barney and Clark, 2007;
performance, i.e. theories that can be used to ex
Mahoney and Pandian, 1992), however, this is
ante assess the performance, and subsequently
difficult to measure ex ante. For the purpose of
the value, of a USO. There are especially two
this study, and in the context of valuation, the
such theories that have received notable
closest proxy is then whether the resource is
attention, namely the resource-based perspective,
available to the firm or not. These reflections
which views a firm as a bundle of resources that
lead us to develop the conceptual framework
ultimately drives performance (Penrose, 1959),
presented in figure 1, where resources are used
4 Arif M. Mirza, Ole Rønning

as valuation inputs rather than financial data. capital resource pool and the circumstances
Whilst traditional valuation methods emphasize under which these resources function. Lastly,
the use of financial returns as inputs, our Wright et al. (1994) argue that human resources
conceptual framework leverages the fundamental are non-substitutable because they are
drivers of these returns, namely resources. generalizable, i.e. can constantly be developed to
Moreover, resources are ex ante measurements ensure that they do not become obsolete, and
which can be used in even the early stages of a therefore the only resources that can substitute
USO, whilst financial returns are ex post them are themselves rare, valuable, inimitable,
measurements and therefore more applicable to and non-substitutable.
mature ventures. As can be seen, the
In the context of new technology ventures,
performance of the firm, and subsequently the
human capital in terms of entrepreneurial roles
financial returns, are also dependent on the
and capabilities is generally highlighted as a key
company's capabilities to utilize the resources.
resource (Colombo and Grilli, 2010; Newbert,
However, this input is not available ex ante.
2007), and in early stage USOs, which often
To test the conceptual framework empirically, have scarce initial resources, human capital in
we present in the following resource- terms of founders and their experience is one of
performance relationships in USOs identified by the most important resources the new venture
previous researchers, and use this to hypothesize has (Shane and Stuart, 2002; Shane, 2004;
relationships between initial resource Colombo and Piva, 2012). Particularly inventor
endowments, and the long term equity value of a presence and the diversity in the founding teams
firm. We further on argue that resources with a background and experience are highlighted as
proven relationship to a USOs long term equity two important human capital factors impacting
value can be used as heuristics in the early stages firm performance.
of a USO to predict its early stage value. We
USOs are usually formed around a technology at
build upon the resource-based theory presented
embryonic stages (Agarwal and Shah, 2014), and
by Mirza and Rønning (2015), and structure our
knowledge to develop, modify or tailor the
discussion in accordance with (Mustar et al.,
technology and associated products or services to
2006), who find that the four resource categories
meet customer requirements is essential for
human, social, financial, and, technology, are
startup success (Di Gregorio and Shane, 2003;
especially important for research-based USOs.
Zucker et al., 1998; Knockaert et al., 2011). This
knowledge is highly tacit, and is typically
Human Resources embodied in the technology inventor (Clarysse et
Wright et al. (1994) write that human capital can al., 2007; Markman et al., 2008; Wright et al.,
be valuable, rare, inimitable and non- 1994). Due to the difficulty in communicating
substitutable, and therefore be the source of tacit knowledge (Polanyi, 2012), the inventor is
sustained competitive advantages. Building on an important asset to exploit the technology, and
the work of Steffy and Maurer (1988), who state numerous studies have highlighted the
that both the demand of labor and the supply of importance of inventor presence in the new
labor is heterogeneous, i.e. that companies venture founding team to achieve high firm
require different skills, and that individuals differ performance (Olofsson and Wahlbin, 1984;
in the skills they possess, they argue that Roberts and Hauptman, 1986; Hess and
individuals contribute differently to a firm, and Rothaermel, 2012; Zucker et al., 2002). Olofsson
human capital can therefore create value. It is and Wahlbin (1984) and Djokovic and Souitaris
also generally a consensus amongst researchers (2008) find that the USOs with the highest
that higher quality human resources lead to growth rates are the companies that involve
higher financial value for firms (Boudreau, 1983; academics who leave the university, and Roberts
Boudreau and Berger, 1985). Wright et al. and Hauptman (1986) argue that the advantage
(1994) go on to write that human capital can of keeping academics close to the new venture is
indeed be rare, if one proxies the quality of due to increased effectiveness of the technology
human capital by cognitive abilities, and the fact transfer process. This is supported by several
that these cognitive abilities are normally scholars (Hess and Rothaermel, 2012; Zucker et
distributed in the population (Jensen, 1980). al., 2002). Knockaert et al. (2010) found that in
Inimitability arises due to unique historical USOs, where the majority of the initial
conditions, causal ambiguity and social researchers became a part of the founding team,
complexity, which complicate the process of the tacit knowledge was transferred effectively
duplicating relevant components of the human and at a sufficient speed, which resulted in
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 5

higher valuations of the firm shares. This leads equity valuations in the long term
them to conclude that the greater the proportion than those that do not.
of the research team that joins the USO as
However important the presence of the
founders, the greater the performance. Based on
technology inventors, spinning off a company
these findings we hypothesize:
with people exclusively from the parent
H1a: Ceteris paribus, USOs who in university can have negative effects on
their early stages have the performance, and an effective combination of
technology inventors on the top management team with both business and
founding team achieve higher long engineering knowledge and experience, is a
term equity valuations than those common denominator for USO performance and
that do not. successful ventures (Gurdon and Samsom, 2010;
Roberts, 1991; Doutriaux and Barker, 1995;
The performance impact of inventors in founding
Visintin and Pittino, 2014; Heirman and
teams is further found to be affected by how
Clarysse, 2004). Shane (2004) writes that it is
renowned the inventor is. USOs that are founded
important for the founding team to have
on the research of a “star” researcher or have a
knowledge about business, management, product
“star” on the founding team can benefit from the
development, production and markets. Grandi
researcher’s reputation. Stars have a strong
and Grimaldi (2005) conclude that successful
ability to send a credible signal about
academic ventures are dependent on founding
unobserved quality of the new venture to
teams that combine individuals with different
external actors (Spence and Michael, 1974).
attitudes, which increases interaction and
Fuller and Rothaermel (2012), examining 238
communication within the founding team, as
USOs in the United States, find that new
well as the allocation of work tasks based on
ventures with star scientists are more likely to
personal attitudes. Further, they write that
achieve IPOs, and Higgins et al. (2011) find that
academic entrepreneurs should devote time to
USOs with star researchers achieve higher IPO
choosing team members that have different
valuations. Zucker et al. (1998), examining the
characteristics and are able to take on different
biotechnology sector, show that star scientists
roles (Grandi and Grimaldi, 2005). Thus, at least
have a direct effect on time to IPO and amount
in the startup phase, it is vital to create
of money obtained at IPO. These findings
differentiated team structures where members
indicate that the renownedness of the inventor
have both research and business profiles
within her field is related to the USOs
(Visintin and Pittino, 2014) because the team is
performance, and subsequently the long term
then better able to adjust to the complex
value of the firm. This is primarily due to three
environmental challenges with which they are
reasons: first, the more renowned the inventor is
confronted (Wright et al., 2012). Knockaert et al.
within her field, the more likely it is that the
(2011) find that it is important for USOs to also
research underlying the technology is cutting
have a commercial mindset to be alert to external
edge and of high quality. Second, more
market movements, and they conclude that
renowned inventors have more to lose if they are
incorporating knowledge about the technology
unsuccessful in starting a new venture due to the
and a commercial mindset in the USOs founding
reputational impacts. Therefore, the fact that she
team leads to enhanced performance. It logically
is willing to bet on the technology sends a strong
follows then, that USOs that are able to diversify
signal about its quality (Fuller and Rothaermel,
in educational backgrounds and previous work
2012), and should ease the process of obtaining
experiences in the founding team should perform
funding and establishing partnerships and
better, and subsequently achieve higher
alliances, which ultimately should increase firm
valuations than those that do not.
growth. Third, the more renowned the inventor
is, the easier it should be to attract additional Being more specific, Rothaermel et al. (2007)
human resources, particularly technological write that experience, capabilities, and
human resources because of aspiring talent knowledge from the industry are critical factors
wishing to work with her. This leads us to the to the success of a spin-off, yet many, if not
following hypothesis: most, management teams lack these capabilities,
negatively affecting their ability to recruit new
H1b: Ceteris paribus, USOs who in
employees and attract early stage capital
their early stages have more
(Clarysse and Moray, 2004; O'Shea et al., 2005;
renowned inventors on the
Hayter, 2013). In agreement, Shane and Stuart
founding team achieve higher
(2002) find that industry experience is
6 Arif M. Mirza, Ole Rønning

significantly related to time-to-IPO. Finally, Walter et al. (2006), examining the impact of
Criaco et al. (2013) show that founders with USOs ability to develop and utilize inter-
entrepreneurial education and university organizational relationships, highlight
experience in terms of research and teaching relationships to research institutions, legal
positively affect firm survival. We hypothesize: authorities, customers, and suppliers as
particularly important to attain growth due to
H2: Ceteris paribus, USOs who in
lacking internal markets and industry knowledge
their early stages have founding
amongst founders. They argue that these
teams that are more heterogeneous
relationships allow the USO to target a larger
in educational backgrounds and
market in less time, learn about customers to
work experiences achieve higher
develop marketable offerings, and ensure timely
equity valuations in the long term
and state-of-the-art resources. Followingly they
than those with more homogeneous
find that sales growth, profit attainment, realized
founding teams.
competitive advantages, and long-term survival
are influenced by a USOs ability to develop
Social Resources external relationships
It is generally agreed upon that a high level of Ties to the parent university are particularly
social capital, in terms of personal relationships, highlighted in previous research as important for
often assist in providing access to venture USO performance. Rothaermel and Thursby
capitalists, business information and potential (2005) write that USOs with strong ties to their
customers (Florin et al., 2003), and the social parent organization are less likely to fail.
capital of founders has therefore been noted in Colombo and Grilli (2010) find that universities
entrepreneurship literature to have several influence the growth rates of local USOs, and
positive impacts on new ventures. Social Bonardo et al. (2011) state that for firms that
resources can be valuable because they can have chosen to go public, the ones that publicize
provide access to information and advice (Hoang that they are a university spinoff experience that
and Antoncic, 2003), key talent and market their origins are beneficial, and that university
information (Freeman, 1999), and as Brush et al. affiliation is positively recognized by investors
(2001) writes, social resources can be used to through enhanced valuations. University
create an image of success that can be leveraged affiliation also enhances the acquisition
to obtain other benefits, such as cooperation and attractiveness to other companies, and the
trust, financial resources, or assets and advantage of the affiliation is correlated with the
equipment purchased at less expensive prices. presence of academics in the top management
Social resources can for obvious reasons be rare team (Bonardo et al., 2011). A strong affiliation
as relationships to more prominent entities are between the USO and its parent university is
more difficult to obtain. Inimitability is likely further likely to be valuable because the parent
due to relationships being built over time and can provide access to resources such as talent
being path dependent (Zander and Zander, 2005; and critical financial resources (Bigliardi et al.,
Santala and Parvinen, 2007). Finally, social 2013), knowledge and complementary R&D
resources can be non-substitutable due to each (Lubik et al., 2013), labs, and expensive
entity in the relationship being unique in terms of equipment at low cost (Starr and MacMillan,
experiences, resources, and networks. 1990; Roberts and Malonet, 1996). We therefore
In entrepreneurship literature, social ties of hypothesize:
founders are shown to reduce informational H3a: Ceteris paribus, USOs who in
asymmetries between founders and potential their early stages have strong ties
investors and partners (Shane and Cable, 2002), to the parent university achieve
and resources made available through higher long term equity valuations
entrepreneurial networks are shown to greatly than those with weak ties.
enhance the survival and growth of new firms
(Brüderl et al., 1992). In new high-tech firms, The performance effect of having strong ties to
who are generally believed to have higher the parent university is further found to be
informational asymmetries, social ties are dependent on the quality of the university.
especially acknowledged to be critically Colombo and Grilli (2010) state that the quality
important (Bidault and Cummings, 1994; Stuart, of the research conducted at the university has a
2000). positive impact on the growth rate of the spin
off, establishing a theoretical link between
university quality and USO performance. Powers
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 7

and McDougall (2005) write that access to the annual employment growth of the new
people with expert knowledge and talent is a venture, and conclude that initial capital invested
critical human capital resource needed for the in USOs is a major predictor for early growth.
development of cutting-edge technology, which Further on, initial capital invested in USOs is
are more likely to be present in more renowned tied to the ability to raise capital later in the
universities, and finds that faculty quality startup process, which is identified as the single
impacts the number of USOs from the university most important factor for achieving an IPO
that go public. Further on, Di Gregorio and (Hayter, 2013; Clarysse et al., 2011). We
Shane (2003) suggest that it is easier for followingly propose:
academics at high ranked universities to get
H4: Ceteris paribus, USOs that
funding for their startups due to their increased
receive investor capital in their
credibility. In support, O'Shea et al. (2005)
early stages achieve higher equity
conclude that the presence of top ranked science
valuations in the long term than
and engineering students and faculty is positively
those that do not.
correlated with USO performance. We therefore
hypothesize: It is important to note here the inherent reverse
causality of capital on firm valuation. It is
H3b: Ceteris paribus, USOs
obvious that raising capital will increase the
originating from high quality
value of a firm equivalent to the amount of
universities achieve higher equity
capital raised, known as pre-money and post-
valuations in the long term than
money valuations. The authors however argue
those originating from lower
that this is not equivalent to increasing the long-
quality universities.
term value of the firm, where the actual
utilization of the capital will be the determining
Financial Resources factor, which is the relationship this study
Due to the early stages of the technologies on investigates. Reverse causality followingly does
which many USOs are founded, the development not bias the hypothesis.
process is long (Rasmussen et al., 2013) and
resource demanding (Shane, 2004; Clarysse et Technical Resources
al., 2011), and access to financial resources is
Shane and Stuart (2002), following Merges and
typically seen as the most important factor for
Nelson (1990), argue that the strength of a
USO growth (Hellman and Puri, 2000; Heirman
USO’s technological endowments at founding is
and Clarysse, 2004; Shane and Stuart, 2002;
an important predictor of its subsequent
Clarysse et al., 2011). Financial resources are
performance because these companies do not
obviously valuable, as they enable the firm to
have complementary assets in place. From a
acquire necessary resources, and can also be
venture capitalist standpoint, Knockaert et al.
rare, because the demand is larger than the
(2006) show that strong technology endowments
supply. However, financial resources are neither
and intellectual property protection are valued
inimitable nor non-substitutable, and can
when deciding to invest in high-tech startups.
therefore not on their own be a source of
Pérez Pérez and Sánchez (2003) summarize by
achieving sustained competitive advantages.
stating that at least initially, USO success is
They do however enable the firm to acquire
more dependent on technological development
other resources that are VRIN, and can therefore
(i.e. strong technological resources) than
from a resource-based perspective indirectly be
marketing, sales, and distribution.
tied to performance.
From a resource-based perspective,
Gurdon and Samsom (2010), interviewing 22 US
technological resources can indeed lead to
and Canadian USOs find that access to capital
sustained competitive advantages. The resources
was a common denominator amongst the
can be valuable as they enable a firm to conceive
successful ventures, and Hayter (2013) argues
of or implement strategies that improve its
that USOs that receive venture capital have a
efficiency and effectiveness (Barney, 1991).
20% to 26% higher likelihood of
They can be rare in terms of the resources being
commercialization success compared with USOs
uniquely owned by the firm, through for example
that do not. In support, Zerbinati et al. (2012),
intellectual property rights. Inimitable because of
measuring performance in terms of employment
intellectual property rights, or in the case of
growth, find that the higher the amount of capital
physical technology, the exploitation may
a USO attracts in its early stages, the higher is
involve socially complex resources where only
8 Arif M. Mirza, Ole Rønning

Figure 2. Conceptual framework

firms that possess the necessary culture, performance, arguing that that the larger the
traditions and social relations can fully exploit number of patents held by a USO at the time of
the technology (Wilkins, 1989). Further on, founding, the smaller the chance of failure.
Narayanan (2000) characterize the knowledge Kamiyama and Sheehan (2006) see that firms
underlying the endowed technology along three increasingly are exploiting their patents as a
dimensions: scope, newness, and tacitness. The means of tapping into external sources of
latter of these, i.e. a high degree of tacitness, can financing. Ownership of patents can demonstrate
make imitation difficult. The two former can to potential investors that a small firm has a
lead to non-substitutability, where broad novel invention with which it may be able to
technological scopes in terms of platform differentiate its products or services from those
applications, and newness in terms of the degree of its competitors, as well as the legal means to
with which the innovation departures from prevent competitors from implementing their
existing technologies (Bierly et al., 2009), can invention in the marketplace. Supporting this,
make the technology difficult to substitute. Clarysse et al. (2007) find that the perceived
quality of the endowed technology of the USO,
Investigating patents as technological resources,
measured in number of patents, enable larger
previous research indicates they can be a
amounts of startup capital to be obtained. Thus,
valuable organizational resource for competitive
because patents can play a large role in enabling
advantage, and predictive of firm performance
firms to attract venture capital investments, they
(e.g., Deeds et al. (2000), Zahra and Bogner
can also affect firm performance, as funding is
(2000), Powers and McDougall (2005)). A
seen as an important factor for achieving new
rational inventor is only likely to patent an
venture success (Zerbinati et al., 2012; Hellman
invention when the expected payoff from
and Puri, 2000). Therefore we propose:
holding the patent outways the cost of obtaining
it (Long, 2002). Followingly, patents have H5: Ceteris Paribus, USOs with
important signaling effects towards external more patents in their early stages
resource holders, where more patents signal achieve higher equity valuations in
higher expected returns. Further on, Hsu and the long term than those with less
Ziedonis (2008) for example finds that patents patents.
act as signals towards investors for the venture’s
The above discussion can be summarized in the
quality when they need it most, namely in the
conceptual framework presented in figure 2.
early stages of development when uncertainty is
at its largest.
In the context of USOs, Shane and Stuart (2002)
establish a linkage between patents and
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 9

Research Methodology information we used Proff1 to retrieve financial

information, Google Patents for patent
To test the developed hypotheses we employ a information, Google Citations for citation rates,
cross-sectional quantitative method. More and LinkedIn as well as university websites for
specifically, we use multiple linear regression. information about founders.
The sample of 63 firms was identified through a
Data Sources and Case Selection funneling process. First, because the EntPro
The sample we analyze comprises of 63 firms database covers the entire population of science-
from the Enter research project which uses the based firms, all firms not originating at a
Norwegian FORNY program as its empirical university and not adhering to the definition of
context. Established in 1995, FORNY is a USOs used in this paper were excluded.
program initiated by The Research Council of Thereafter, we limited our sample to companies
Norway to commercialize results from publicly established between 2000 and 2005. This is due
financed research institutions (Rasmussen et al., to data being insufficient for companies
2013), with a general objective being to increase established in 1999 or earlier, and research
wealth creation in Norway (Borlaug and STEP, indicating that USOs take an average of 10 years
2009). EntPro is a project initiated to follow the to commercialize their technology (Rasmussen et
development of the companies in the FORNY al., 2013; Lawton Smith and Ho, 2006). By
portfolio, which as of mid 2013 consisted of 474 limiting our sample to companies that have been
science-based companies (Rasmussen et al., active at least a certain number of years, we limit
2013). Due to financial incentives for registering the possible bias from including companies that
a firm in the program as well as legislative have not come far enough in the
change in Norway in 2003 that, similar to the commercialization process to be valued properly,
Bayh-Dole Act of 1980 in the US, grants and therefore receive erroneous valuations.
intellectual property rights to the parent After this first step, a total of 72 firms were left
institution instead of the individual researcher, it in the sample. To retrieve and structure relevant
is reasonable to say that the EntPro database data for our specific research, a codebook was
covers the majority of the population of science- developed where all relevant variables where
based firms in Norway established after 2000. operationalized into objective measures. This
This eliminates sampling biases in our study resulted in a total of 65 variables, and the final
which are common in entrepreneurial research variable codebook can be found in the appendix2.
(Short et al., 2010; Mullen et al., 2009). The The coding scheme was reviewed with a senior
database consists of business plans, corporate researcher3 at NTNU to ensure its validity. Each
announcements from the Norwegian business USO was coded by a single researcher on the
registry, annual reports, news articles, survey team to ensure internal coding reliability in each
data and context-level data on institutional observation, and subsequently saved in a master
factors. Further on, this data has for the majority template containing all the USOs. To reduce
of the USOs in the database been structured and possible bias from not having several researchers
coded into a standardized template by a prior assess the same venture, the researchers together
research group, as a part of a larger research developed a set of coding criteria for each
project between NTNU and Bodø Business variable. The coding was conducted in two
School. A great strength in using the EntPro phases. First, information in the EntPro database
database as a primary source of data is that was reviewed. In phase two, the missing
companies that are no longer in official registers information was retrieved through additional
are still in the EntPro database, which eliminates information sources such as previously
survival bias in our sample. mentioned Proff, Google Patents, Google
Seen on an international scale, the EntPro Citations and LinkedIn.
database provides a vast amount of unique data Upon completion of coding the 72 USOs in our
on the USOs in our sample. To supplement this, sample, 9 USOs were excluded. Five companies
additional data was collected through online were excluded due to difficulties in estimating
sources. For example, information about the their values. This was because the companies at
founding team was missing in many business
plans, patent information was rarely mentioned,
1
and the latest annual reports where sometimes Proff.no is a leading provider of in-debt financial data
not available. To retrieve the missing on Norwegian registered firms.
2
Appendix 3
3
PhD-candidate Marius Tuft Mathisen
10 Arif M. Mirza, Ole Rønning

the time of measurement had not completed their are several limitations to our measure of firm
product development, and subsequently had not value, which we return to later in the paper, a
conducted their first commercial sale. Because great strength is its simplicity. A limited amount
these companies had no or low revenues, of financial data is necessary to apply the
including these companies in the sample would method, and it implicitly forecasts future
have resulted in serious valuation errors. Four expected cash flows (Baker and Ruback, 1999).
companies were excluded due to changes in Moreover, valuing a mature company using a
organizational structures that made it impossible multiple of an operating or financial measure is a
to code the original firm and its resources. This common and popular approach. Using third-
left a final sample of 63 firms. party valuations on the other hand, which for
private firms are often affected by contractual
Variable Measurements terms such as vesting of capital and differences
in equity claims (Damodaran, 2009), would have
When employing a quantitative method, it is biased our dependent variable as we would not
necessary to identify relevant variables, relate be able to identify, and followingly adjust values
those variables to the hypothesis in question, and for, such terms. Lastly it would be difficult to
finally use valid and reliable measurements to obtain up-to-date valuations on each firm. These
conduct the analysis (Creswell, 2003). There is limitations restricted our valuation options to
no single way to operationalize the variables relative valuation.
used in our analysis, however, as far as the
availability of data allows it, we follow the Mirza and Rønning (2015) outline the process of
established definitions employed in existing valuing a firm using relative valuation. In this
research, and acknowledge limitations when that study we use industry multiples to average out
is not possible. the errors that would occur if a multiple from a
single comparable firm was used instead. The
Dependent Variable use of industry classifications to identify
multiples is a common practice in valuation
This study investigates the relationship between research (Liu et al., 2002). The question
initial resources and firm value, and the regarding which specific multiple to use has
dependent variable is followingly firm value. received a lot of attention in corporate finance,
Several limitations, therein access to valuation and there seems to be consensus that earnings-
data as well as the validity of valuations multiples perform the best (Liu et al., 2002).
conducted by third parties which are often However, Damodaran (2012) writes that sales-
dependent on contractual terms (Damodaran, multiples can be more appropriate when the firm
2012), led us to firsthand value each company in has negative earnings. Further on, many
our sample. Amongst the three most common valuation studies focus on enterprise multiples as
valuation methods, the cash flow discounting they ignore the effects of different capital
method requires a greater insight into the structures by valuing both equity and debt (Deng
company's financials and future growth than et al., 2012). Because the companies in our
what was possible for us to retrieve, whilst sample are likely to vary in capital structures, we
valuing the firms using a balance sheet approach use enterprise value-multiples. Moreover, we use
would not have included the future profit two different scaling factors: in accordance with
potential in these firms (Mirza and Rønning, the established acceptance that earnings-
2015). We therefore opted to use the income multiples perform better we use EBITDA as a
statement method relative valuation, also known scaling factor on companies that show stable
as the multiples approach. Mirza and Rønning earnings without large extraordinary
(2015) find that relative valuation is not suitable development expenses. If a company does
for early-stage firms. However, as we are display such extraordinary expenses, using
assessing the firm values at more mature stages EBITDA will not be representative of how the
(companies in our sample are founded during or company is likely to do financially when the
prior to 2005, and the majority of firm values are development costs are removed, and using
calculated using data from 2013), one of the EBITDA as a scaling factor will subsequently
major obstacles of lacking financial outputs is result in large valuation errors. In accordance
removed. Further on, we removed from our with Damodaran (2012), we use the EV/sales-
sample firms that at the time of measurement had multiple on these companies instead. A more
not conducted their first commercial sale to objective approach would have been to
eliminate companies with unrepresentative consistently use one specific multiple, or an
revenues of normal operations. Although there average of the two, on all companies. However,
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 11

valuation is a subjective process, and being Step 4: As previously mentioned, using

rigorous on this would likely have resulted in enterprise-multiples to value a firm results in the
more inaccurate valuations because of the lack of enterprise value, which Liu et al. (2002) defines
customization. Our process of valuing each firm as:
can be divided into five steps: !"#$%&%'($)*+,-$ =
Step 1: First, each company was classified as /+%0$#)*+,-$)12)$3-'#4
either unsuccessful, still operating, acquired or
completed IPO. None of the firms in our sample +)6110)*+,-$)12)7$6#
had completed an IPO, and this category was −)(:+(ℎ)+"7):+(ℎ)$3-'*+,$"#())
therefore removed.
Step 2: To retrieve relevant industry multiples,
Because we are only interested in the return to a
each company in each category was mapped to
potential investor, the market value of equity for
two industries using industry classifications4.
each firm was calculated by subtracting debt in
Two data sources with different industry
the company from the enterprise value, and
classifications were used due to difficulties in
subsequently adding in cash and cash
placing some of the firms in our sample using
equivalents.
only one of the classifications. To assure a good
industry match, other companies in the chosen =+,-$)12)$3-'#4 =
industry were assessed to assure that the core !"#$%&%'($)*+,-$
activities matched those of the USO being
−)6110)*+,-$)12)7$6#
valued.
+)(:+(ℎ)+"7):+(ℎ)$3-'*+,$"#()
Step 3: Each company in each category was
thereafter valued at time T, where T represents )
the earliest point in time an investor could have Step 5: Because we used public multiples to
exited their investment in the USO. For acquired value the majority of firms in our sample, an
companies, T represents the year the company illiquidity discount was applied. Compared to
was acquired. For still operating firms, T public firms, investments in private firms are
represents the year of the latest available annual illiquid because of fewer buyers/sellers and
report. substantial transaction costs when conducting a
Table 1 summarizes the valuation method for buy/sale (Damodaran, 2005). Followingly, when
each category of firms. valuing equity in a private company based on
public company multiples, one should include an
Company(Status Definition Valuation(Method illiquidity discount rate. Although illiquidity
Unsuccessful 1.+Company+is+removed+from+ Firm+value+set+to+0 discounts often are applied as a fixed rate
official+business+registers+and+
there+is+no+information+
between 25-35%, Damodaran (2005) argues that
indicating+a+merger+or+ they are highly firm specific, and although there
aquisition.
is no easy way to determine this discount,
2.+Company+has+abondoned+ illiquidity discounts should be adjusted to each
original+business+idea firm. Silber (1991) developed an equation5 to
Still+operating Annual+report+showing+activity+ Valued+using+EV/EBITDA+or+EV/Sales.
calculate the size of the liquidity discount by
and+the+company+has+
conducted+their+first+ analyzing the valuation difference between a
commercial+sale company's public and restricted stock. He finds
Acquired 100%+of+the+company's+shares+ Valued+equal+to+the+acquisition+
have+been+bought+by+another+ amount,+if+available.+If+not,+valued+
that the illiquidity discount is dependent on the
company using+acquisition+or+IPO+multiples+ firm’s revenues, the proportion of restricted
(Price/sales)+from+similar+companies+in+
terms+of+products+or+services+being+
stock to publically traded stock, if the company
sold.+The+use+of+multiples+to+value+and+ has positive earnings, and if there is a customer
acquisition+target+is+used+in+more+than+ relationship between the investor and the firm.
50%+of+all+acquisitions+(Damodaran,+
1996). Although Silber (1991) based his analysis on
public firms using restricted stock, Damodaran
Table 1. Valuation method for each category of firms
(2005) points outs that he does provide a
roadmap for how to calculate private firm
4
Professor Damodaran at Stern University
discounts. For the lack of better ways, an
(http://pages.stern.nyu.edu/~adamodar/), as well as illiquidity discount using Silber (1991) equation
Ycharts, an online provider of financial data. Professor
Damodaran lists 95 industries, whilst the second data
5
source lists 148. Appendix 7
12 Arif M. Mirza, Ole Rønning

was followingly calculated for each firm in our Inventor Renownedness (H1b)
sample and applied to the equity value calculated
Fuller and Rothaermel (2012) proxy the presence
in the previous step. The final valuation equation
of a star inventor by her inclusion in the top one
used in this study can be summarized as
half of one percent of cited scholars in her field.
following:
A citation represents evidence that the person
!3-'#4)*+,-$' being cited has done work that is viewed as
=' = )
>,,'3-'7'#4)7'(:1-"#' relevant to the current research frontier (Jr,
1986). Cole and Cole (1967) argue that citations
!3-'#4)*+,-$? =)
can be viewed as a form of recognition, and
=+,-#'1")@-,#'&,$)×)B:+,'"C)2+:#1% Diamond (1984) that citations can be used as a
− D$6# + (E+(ℎ)+"7):+(ℎ)$3-'*+,$"#() proxy for a researcher's ability to do quality
where research. Following this, we use the total number
of citations on any article, authored or co-
Valuation)multiple = The multiple used: either authored by the firm founders, as a proxy for
EV/EBITDA, EV/Sales or, if acquisition, how renowned the inventor is. All founders are
P/Sales. included due to difficulties in separating
Scaling)factor = EBITDA or sales in year T for the inventors from founders in many of the case
firm being valued. firms. For each founder, a “My citations” profile
was created in Google Scholar. All articles
Debti = Book value of firm i’s debt in year T authored or co-authored by the founders were
(Cash)&)Cash)equivalents)i = Cash and cash added to their respective profiles. After adding
equivalents held by firm i in year T all articles, each Google Scholar My citations-
profile displayed a bar graph with year on the x-
Independent Variables axis and the number of citations received that
year on the y-axis. For all founders in the firm,
The independent variables in our study consist of all citations up to and including the second year
the resources with which we have hypothesized a of operations were summed together to a total
relationship to firm value. The variables are citation count.
measured through proxies, and because we are
interested in early-stage predictors of a USOs Degree of Heterogeneity in Founding Team
value, they are measured at the end of the second (H2)
year of operations. The second year, instead of
the first, is chosen to account for differences in Following literature on managerial work
time of founding during the first calendar year. behavior and key managerial experiences,
Measuring the variables at the end of the second backgrounds and skills needed for high
year also limits biases that might occur when performance (Katz, 1974; Kotter, 1982; Whitley,
companies are registered prior to establishing a 1989; Hambrick et al., 1996) we focus on
proper organization, and that are likely to common tier 1 positions as well as educational
experience changes in management during the backgrounds and work experiences highlighted
first couple of months. Further on, founding as important for performance in USO literature.
team members are defined as individuals with Founding team heterogeneity is followingly
equity in the new venture. defined in terms of educational backgrounds and
work experiences from the industry (Roberts,
Inventor on Founding Team (H1a) 1991; Rothaermel et al., 2007; Shane and Stuart,
2002), technical experience (Di Gregorio and
We use a dummy variable to indicate whether Shane, 2003; Zucker et al., 1998; Knockaert et
the technology inventor is present in the al., 2011), managerial experience (Szilagyi and
founding team. The variable is coded 1 if at least Schweiger, 1984), entrepreneurial experience
one (if there are several) inventor is present in (Criaco et al., 2013; Brüderl et al., 1992),
the founding team. This was coded by first financial experience and marketing experience.
identifying the inventor name(s) using the
ventures business plan, and subsequently A common measure of experience in existing
reviewing annual reports to identify if the streams of research is to use the number of years
inventor had an equity stake in the company. of experience the founders have in the field of
interest. However, data limitations prohibited us
from attaining such detailed information. Instead
we follow Shane and Stuart (2002) and Miloud
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 13

et al. (2012), and use dummy variables. ranking published by Webometrics6, the largest
Technological experience is coded 1 if any of the academic ranking of Higher Education
founders have educational backgrounds or work Institutions with regards to number of
experience related to the technology of the institutions, and the only ranking that covered all
company. Industry experience, marketing the universities in our sample. Webometrics uses
experience and finance experience are coded 1 if link analysis based on a composite indicator that
any of the founders have work experience or an weighs the volume (number of files and pages)
educational background in marketing or of web content, and the impact and visibility of
finances, respectively. Top management the university's publications in terms of external
experience is coded 1 if any of the founders have site citations, to evaluate university quality. A
previously held a position in the top two tiers of low rank indicates high quality, and to avoid
an organization (Roure and Keeley, 1990; reversed signs in our model, we inverted the rank
Carpenter and Fredrickson, 2001). That includes through the following transformation: 1000/X
positions such as CEO, CFO, CTO, President or (1000 instead of 1 is used in the numerator to
Senior Vice President etc. Finally, startup avoid small fractions being rounded to 0).
experience is coded 1 if any of the founders have
previously started a company. Because we have Early-Stage Investor Capital (H4)
no theoretical foundation to argue the relative
Using the data available we were able to identify
importance of each experience, we make the
firms that have received investor capital,
assumption that they are equally important, and
however we were only able retrieve the actual
define the degree of heterogeneity as the sum of
investment amounts in 7 of the firms in our
the dummy variables. To conduct the coding,
sample. Instead of using the actual amount we
information about founders educational
therefore use the crude proxy of a dummy
backgrounds and work experiences was first
variable that is coded 1 if the company has
retrieved from the company's business plans.
received investor capital during the first two
Missing information was subsequently collected
years.
from LinkedIn.
Number of Patents (H5)
Ties to The Parent University (H3a)
Calculated as the total number of patents filed or
There is no single and perfect way to measure
published during or prior to the first two years of
the strength of a tie, however, a common and
operations where the USO is listed as assignee.
simple proxy for the strength of a social tie is the
To identify relevant patents, business plans were
number of interactions between two people in a
first reviewed for each company. Thereafter, a
given amount of time (Nelson, 1989).
search on the company name, as well as the
Granovetter (1973) expands on this and defines
inventors, was conducted using Googles patent
the strength of an interpersonal tie as the linear
search engine. Only patents that were filed or
combination between the amount of time, the
published during or prior to the second year of
emotional intensity, the intimacy and the
the firm’s operations, and where the firm itself
reciprocal services that characterize each tie.
was listed as assignee, were included in our
Because such detailed information about the ties
results.
between the venture and its parent organization
is unavailable to us, we instead use the total
number of years the founders have spent at the
Control Variables
parent university as a crude proxy. That includes There are numerous factors, other than those of
time spent as employee, researcher or PHD- primary interest, that can affect the valuation of a
candidate. The rationale being that the longer company, leading to inflation in the error
history a founder has with the parent university, variance in our model (Mullen et al., 2009). To
the stronger the founders network is likely to be, control for these factors, and reduce this error,
and the more the founder is able to draw upon we incorporate control variables. In addition,
the university's resources. every independent variable in the model acts as a
control for the other variables.
Quality of Parent University (H3b)
Following Di Gregorio and Shane (2003), who
6
proxy university eminence through its ranking http://www.webometrics.info/en/Europe/Norway - an
published in the Gourman Reports, we proxy the initiative by the Cybermetrics lab belonging to the
quality of the university through its world Consejo Superior de investigaciones Cientificas, the
largest public research body in Spain.
14 Arif M. Mirza, Ole Rønning

Previous research has highlighted several factors Variable Sign Measurement

that influence valuation. For example, Gompers
Company(value Dependent EV/EBITDA,(EV/Sales(or(Price/Sales(multiples
and Lerner (1999) find a strong correlation
Dummy(variable(with(‘1’(indicating(the(
between capital infusion into the public market, Inventor(on(founding(team + technology(inventor(being(on(the(founding(
measured by public stock market indexes, and team
Total(number(of(citations(on(any(article(
valuations in the venture capital market. Market Inventor(renownedness + authored(or(coGauthored(by(the(founders(as(
size and industry profitability and return on calculated(by(Google(Scholar.(
Sum(of(dummy(variables(indicating(different(
investment are also highlighted. At the firm- Team(heterogeneity + backgrounds(and(experiences(in(the(founding(
level, the development stage, firm age, and type team
Total(number(of(years(the(founders(have(
of industry in which the firm is situated have Ties(to(parent(university +
worked/researched(at(the(parent(university
traditionally been controlled for. However, a rule
Parent(university(quality + The(university's(world(ranking(by(Webometrics
of thumb in regression is to have at least 10 data
Dummy(variable(with(‘1’(indicating(received(
cases per variable (Field, 2013). As further Investor(capital + investor(capital(during(or(prior(to(the(first(two(
discussed in the limitations section, with our years(of(operations
Total(number(of(patents(published(or(filed(
small sample we are already at the threshold of Number(of(patents +
during(or(prior(to(the(first(two(years(of(
this value. Therefore, we need to limit the Control2Variables
number of control variables in our model. We Year(of(last(annual(report,(year(of(failure(or(
Company(age
omit using the OSEBX stock index to control for year(of(acquisition,(minus(year(of(founding.

capital infusion in the public market as our Table 2. Variable definitions and measurements
valuations are in 95% of the cases conducted in
the same year (2013, which is the year of the Although only normality of residuals is required
most recent annual report available for still in regression, normally distributed variables are
operating companies), and the variable would more likely to produce normally distributed
therefore not account for any significant variance residuals. We therefore checked all the
in the model. It is reasonable to believe that continuous variables for normality, first by
USOs that target large markets, ceteris paribus, conducting a Shapiro-Wilk W-test, and if it was
also receive higher valuations because of the significant, we reviewed the deviation from
larger income potential. We therefore include normality by assessing the skewness and kurtosis
market size as a control variable at the market- of the distribution. All variables showed
level. Market size is based on the size of the significant deviations from normality. Four of
industry in MUSD in which the company was the variables showing the largest deviations,
placed during the valuation process, and is including the dependent variable, were
retrieved from Ychart7. Lastly we include firm transformed to reduce the skewness and kurtosis
age, which is calculated as the difference of their distributions. After trial and error,
between the year of the last available annual ‘Company value’, ‘Degree of Heterogeneity’ and
report, year of failure or year of acquisition, and ‘Inventor renownedness’ was LN-transformed,
year of founding. whilst ‘Number of patents’ was transformed
using [1-1/(1+Number of patents)]. All the
Model Estimation and Robustness variables still showed a significant deviation
from normality after the transformation when
To summarize the previous discussion, table 2
assessed using the Shapiro-Wilk W-test.
presents the variables in our model.
However, a review of the variables skewness and
kurtosis revealed a significant reduction in both
measures for all transformed variables. The
following equation represents the model to be
estimated and analyzed:

7
Ycharts.com is an online provider of financial data.
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 15

," 1 + :1@&+"4)*+,-$ = _ ) did a Shapiro-Wilk-test on the standardized

+`a × ," 1 + b$+@)ℎ$#$%1C$"$'#4 ) residuals which was non-significant (p = 0.395).
+`c × b'$()#1)&+%$"#)-"'*$%('#4
Methodological Limitations
1
+`d ×(1 − There are several limitations to the methodology
1 + e-@6$%)12)&+#$"#(
employed in this study, where the measurement
+`f × >"*$(#1%):+&'#+,
of the dependent variable as well as the sample
+`g × >"*$"#1%)1")21-"7'"C)#$+@ size are the most crucial.
1000 The dependent variable is measured firsthand
+`h ×
j+%$"#)-"'*$%('#4)3-+,'#4 through the use of valuation multiples, and the
+`k × ," 1 + >"*$"#1%)%$"1l"$7"$(( value estimations are followingly limited by the
availability of information and the shortcomings
+`m × E1@&+"4)+C$
of relative valuation when applied on young
+`n ×(/+%0$#)('o$)) ventures. First, we have ourselves valued the
) companies based on the limited information we
had available. This may have led to other
Here, α represents the intercept term, β1 - β7 are valuations than what more informed analysts
the coefficients of the independent variables, and would have estimated. Second, the fundamental
β8 and β9 the coefficients of the control argument underlying relative valuation is that
variables. Prior to conducting the analysis, similar assets should be priced equally (Liu et
variables where controlled for data entry errors, al., 2002). The validity of our valuation estimates
and the few single missing values were, therefore rest on the similarity between the
following Tabachnick and Fidell (1996), private firm being valued and the industry from
replaced with the variable mean. To control for which the multiple is derived. USOs are in many
outliers and cases with undue influence on the cases firms commercializing unique technologies
model we calculated Cook’s Distance, which (Shane, 2004) whose comparability to other
measures the overall influence of a case on the firms in the same industry may be limited (Mirza
model, and the DFBetas, which looks at the and Rønning, 2015). Further on, factors affecting
difference between a parameter estimated using value are likely to be different for private and
all cases, and estimated when one case is public firms, so public multiples are not directly
excluded. A Cook’s distance greater than 1 transferrable to a private company. Third, the
(Cook and Weisberg, 1982), or a DFBeta greater industry multiples used are crude, and in some
than 2 (Stevens, 2012) indicates a case with cases it is difficult to place a firm in the right
undue influence on the model. All Cook’s industry, again because of the unique
distances were well below 1 (highest was 0.176), technologies these firms are founded on.
and the largest DFBeta was 0.86. This indicates Optimally, the industry classifications used
no outliers or cases with undue influence in our would have been more fine grained to easier
model. To assure independence of residuals, or separate firms into different industries. Fourth,
lack of autocorrelation, we used the Durbin- the value of the firms are calculated at a specific
Watson statistic which tests for serial point in time, at which point the firm may or
correlations between residuals. In a model with 9 may not have reached its full commercial
regressors and 65 cases8, a lower bound for the potential in terms of revenues. The potential
statistic is 1.301 and an upper bound 1.923, at 5- future growth in revenues and earnings a firm
percent significance (Durbin and Watson, 1951). might experience is followingly not incorporated
With a value of 1.837 in our model there are no into our value estimates. The dependent variable
indications of autocorrelation. Inspection of also limits the model's generalizability as it is
residual plots revealed a random array of evenly constrained in its measurements, i.e. the value of
dispersed dots around zero which indicates that a company can theoretically speaking be
the variance of the residual terms are constant, unlimited, whilst the cases in our sample only
and that there is no problem with cover a specific interval.
heteroscedasticity. Finally, to check the
assumption of normally distributed errors, we The use of relative valuation was despite the
limitations necessary to conduct this study, and
we have to our best ability addressed these
8 issues. Instead of using multiples from a single
https://www3.nd.edu/~wevans1/econ30331/Durbin_Wats firm we have opted to use industry multiples to
on_tables.pdf
16 Arif M. Mirza, Ole Rønning

average out the error that would be obtained in followingly believed to be a good fit to the
using a single firm in the chosen industry, and variable of interest, however, without estimating
also added an illiquidity discount to account for the actual correlations, there is risk of poor
the effect of using a public multiple on a private proxies resulting in invalid and biased results.
firm. Further on, to cross check that the right Finally, the methodology is limited by the
industry was chosen, other companies in the reliability of the information sources used. The
same industry were reviewed to verify that they business plans and annual reports are considered
delivered the same type of products or services. more reliable than online sources such as
We have also excluded firms from our sample LinkedIn and Google, nonetheless, to retrieve the
that have not completed their first commercial information needed to conduct a study of this
sale to ensure that the firm has revenues from magnitude, a 360 degree approach to gathering
their core activities. In any case, valuing a data from all available sources was deemed
company is an inherently difficult and subjective necessary.
task, and although there are severe limitations to
the absolute measurement of our dependent Results
variable, the methodological limitations are
likely to inflict constant valuation errors across In the following section we present descriptives
the firms in the sample, and the relative value of followed by an inferential discussion on the
each firm should followingly be valid for the hypotheses in question.
purposes of this study.
Descriptives
The second limitation relates to the small sample
size. Field (2013) writes that assuming a 5% Variable Mean Minimum Maximum Std.1Deviation
significance level and the recommended power
Company(value((transformed) 6,884 0,000 20,040 8,360
of a test of 0.8, 783 participants are necessary to
detect a small effect size (Pearson correlation = Human1Resources

0.1), 85 participants to detect a medium effect Inventor(on(founding(team((binary) 0,890 0,000 1,000 0,317

size (Pearson correlation = 0.3) and 28 Inventor(renownedness((transformed) 4,234 0,000 9,720 3,193
participants to detect a large effect size (Pearson
Team(heterogeneity((transformed) 1,057 0,000 1,950 0,411
correlation = 0.5). With 63 participants in our
sample we are unlikely to detect small and Social1Resources

medium-sized effects. The sample size is Ties(to(parent(university((years) 17,400 0,000 54,000 13,710

however limited by the availability of data, and Parent(university(quality((transformed) 5,990 0,150 14,710 4,867
although the sample is small, it does cover the
majority of USOs in Norway established in the Financial1Resources

period 2000-2005, and should be representative Investor(capital((binary) 0,210 0,000 1,000 0,408

of the Norwegian population of USOs. However, Technological1Resources

because the sample is confined to Norwegian
Number(of(patents((transformed) 0,270 0,000 0,990 0,391
USOs, the model's generalizability across
geographies is limited. Control1Variables

Company(age((years) 8,370 1,000 13,000 3,456

Another important limitation in the overall
methodology is the lack of a longitudinal Market(size((MUSD) 429(837 25(081 1(415(745 368(828

approach, i.e. we do not assess how the variables Table 3. Descriptives of the variables in the analysis. See
in question develop over time, but rather assume appendix 7 for descriptives of the untransformed
that the value the USO achieves is due to the variables
presence of initial resource endowments. This is
Our sample consists of 63 firms. Table 3 shows
however necessary as we are interested in the
means, minimum and maximum values for all
explanatory power of these resources ex ante on
variables, as well as the standard deviations. 37
the future value of the firm.
of the firms were categorized as unsuccessful
As for the independent variables, they are and given a value of zero, four firms were
measured through proxies. Per definition, a acquired, with an average valuation of 162
proxy only infers the value of the actual variable MNOK, whilst 22 firms were categorized as still
of interest, and its fit to be used as a proxy is operating with a mean value of 59 MNOK. The
given by the correlation between the two. We mean value across all categories was
have, as far as the available data has allowed us, approximately 31 MNOK, and the average
used proxies that are verified in previous company age at the time of valuation was 8.37
research, or anchored in theory. The proxies are years. The relatively high standard deviation of
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 17

1 2 3 4 5 6 7 8 9 10

Company(value((transformed)([1] 1

Human0Resources

Inventor(on(founding(team([2] 0,2* 1

Inventor(renownedness((transformed)([3] 0,25** 0,299*** 1

Team(heterogeneity((transformed)([4] 0,284** 0,251** 0,144 1

Social0Resources

Ties(to(parent(university([5] 0,401**** 0,229** 0,393**** 0,224** 1

Parent(university(quality((transformed)([6] 0,332*** I0,089 0,186* 0,098 0,246** 1

Financial0Resources

Investor(capital([7] 0,097 0,18* 0,111 0,182* 0,054 I0,036 1

Technological0Resources

Number(of(patents((transformed)([8] 0,391**** 0,047 0,328*** I0,109 0,333*** 0,038 0,079 1

Control0Variables

Company(age([9] 0,494**** 0,185* 0,182* 0,123 0,168* 0,18* 0,335*** 0,323*** 1

Market(size((MUSD)([10] 0,025 I0,155 0,166* I0,08 0,27** 0,324*** I0,001 0,111 I0,027 1

Table 4. Pearson Correlation Matrix - Correlation significant at: * p < 0,1, ** p < ,05, *** p < ,01, **** p < ,001

company age (~3,5) indicates that there is a wide As can be seen, all correlations between the
spread in how long the companies have been dependent variable ‘Company value’ and the
operating which underpins the need to include independent variables are in the predicted
age as a control variable. The average firm direction, and consistent with the developed
values in each of the outcome categories are as hypotheses. Seven out of nine correlations are
expected, i.e. that firms that are acquired are on significant. The only independent variable with
average valued higher than firms that are still an insignificant correlation to ‘Company value’
operating, which are valued higher than firms is ‘Investor capital’. The table also shows that
that have shut down. there are significant correlations between
independent variables, however this does not
The majority of the founding teams had an
necessarily indicate multicollinearity-problems
inventor present (0.89), the founders were on
in the model. A Pearson correlation greater than
average cited 1219 times, and the degree of
0.8 between two variables indicates problems of
heterogeneity was 2.13 out of 6. The latter
collinearity (Field, 2013), and a variance
meaning that on average, firms had founding
inflation factor (VIF) greater than 10 should
teams with backgrounds and experiences from 2
raise concerns (Bowerman and O'Connell, 1990).
out of the 6 fields considered in this study.
Further on, if the average VIF is substantially
Moreover, founding teams had an average of
greater than 1, then the regression may be biased
17.40 years of working experience from the
(Bowerman and O'Connell, 1990). The largest
parent university. Under 25% (0.21) of the firms
Pearson correlation between any two variables is
in the sample received investor capital during or
0.357, between ‘Inventor renownedness’ and
prior to the first two years of operations, and the
‘Ties to parent’. This is somewhat expected as
mean number of patents filed and published was
the longer founders have worked/researched at
3.4 per firm.
the university, the more papers they are likely to
have published and followingly be cited on. The
Inferential
highest VIF was 1.483, with an average VIF of
Table 4 reports the bivariate Pearson correlations 1.260. Despite significant correlations between
between the variables in our model, ordered by independent variables, both the magnitudes of
resource category. the correlations and the VIF confirm that
collinearity is not a problem in our model.
18 Arif M. Mirza, Ole Rønning

In the following we review pure relationships Figure 5 shows the mean value by outcome
between the independent variables in our four category for the financial resource in our study.
resource categories and the dependent variable The graph indicates a positive relationship
represented by the three outcome categories. It is between firm value and attained investor capital,
here assumed, as was shown above, that firms where nearly 20% of unsuccessful firms, nearly
that are categorized as acquired are on average 23% of still operating firms, and 25% of all
valued higher than firms that are still operating, acquired firms received investor capital.
which again are valued higher than firms that are Although indicative, this underpins hypothesis 5.
unsuccessful. It is important to note that the
0,3
below inferentials are only indicative of what
0,25
results may be expected from the regression
analysis. The graphs represent averages, and may 0,2

obscure the real relationship between the two 0,15

variables of interest. Further on, relationships 0,1

between the different independent variables are 0,05
not controlled for. 0
Unsuccessfull Still2Operating Acquired
Figure 3 shows the mean values by outcome
category for the human capital resources Figure 5. Mean value for financial resources (investor
‘Inventor on team’, ‘Degree of heterogeneity’ capital)
and ‘Inventor renownedness’. The graph shows a
positive relationship between the two former Finally, figure 6 shows the mean value by
variables and firm value, i.e. firm value is outcome category for the technical resource in
increasing with increasing degree of our study, i.e. patents. The graph does not
heterogeneity and inventors on the team, which indicate a linear relationship between patents and
indicates support for hypothesis 1a and 2. There firm value, which indicates that hypothesis 6
does not seem to be a linear relationship between may not be supported.
the latter variable and firm value, which 8
indicates that hypothesis 1b may not be 7

supported. This will be further examined in the 6

next section. 5
4

3 3

2,5 2
1
2
0
1,5 Unsuccessfull Still5Operating Acquired
1
0,5 Figure 6. Mean value for technological resources
0 (number of patents)
Unsuccessfull Still2Operating Acquired

Inventor2on2team Inventor2renownedness Degree2of2heterogeneity As mentioned, the above inferentials do not

control for the effect different independent
Figure 3. Mean values for human resources variables assert on each other. Table 5 presents
the complete regression model.
Figure 4 shows the mean values by outcome
category for the social resources in our study. As
can be seen, both the variables ‘Ties to parent’ Discussion
and ‘Parent ranking’ are increasing in firm value,
Because we are analyzing the hypotheses under
which indicates support for hypothesis 3 and 4.
the assumption of ceteris paribus, hypotheses
35 related to each resource category are first tested
30 alone in model 2 through model 5 with only
25
20
control variables. Model 6 is the full model with
15 all independent and control variables included.
10 Because of lacking theory as to the relative
5 importance of each independent variable, all
0
Unsuccessfull Still1Operating Acquired variables are introduced in the full model
Ties1to1parent Parent1ranking1(inversed)
simultaneously. This is in accordance with
researchers believing that this method is the only
Figure 4. Mean values for social resources appropriate method for theory testing
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 19

Variable Model+1 Model+2 Model+3 Model+4 Model+5 Model+6

Human+Resources
Inventor(on(founding(team 0,036 0,067
Inventor(renownedness((transformed) 0,123 ;0,037
Team(heterogeneity((transformed) 0,206* 0,204*
Social+Resources
Ties(to(parent(university 0,312*** 0,182
Parent(university(quality((transformed) 0,222* 0,236**
Financial+Resources
Investor(capital ;0,077 ;0,08
Technological+Resources
Number(of(patents((transformed) 0,257** 0,262**
Control+Variables
Company(age 0,495**** 0,440**** 0,398**** 0,521**** 0,411**** 0,330***
Market(size((MUSD) 0,039 0,039 ;0,120 0,039 0,008 ;0,088

Constant ;3,503 ;9,064** ;5,593 ;3,706 ;2,986 ;10,008***

N((number(of(observations) 63 63 63 63 63 63
Adjusted(R2 0,22 0,255 0,35 0,212 0,268 0,378

Table 5. Regression model - Coefficient significant at: * p < 0,1, ** p < ,05, *** p < ,01, **** p < ,001

(Studenmund, 2011), because stepwise including age as a control variable, as it will

techniques are influenced by random variation in account for the variance this effect has on firm
the data and so seldom give replicable results if value. Market size is surprisingly, and in
the model is re-tested. Model 1 represents the contradiction to existing research, found to have
baseline model, and includes only the control a small and insignificant effect. This may
variables. Given the limitations previously indicate that in the context of USOs, market
mentioned, we focus our discussion on general sizes of existing industries are less important
relationships between initial resource- when valuing a new venture. USOs often
endowments of USOs and firm value instead of develop novel technology (Shane, 2004), and
exact coefficients and effect sizes. this technology might unlock new application
areas and markets, rendering market sizes of the
Model 1 only includes the control variables. Age
industries of competitors less representative for
is estimated to have a large and significant effect
the markets targeted by the USO. Another
at the 0.1% level, and is consistently significant
possible and plausible explanation is that the
across all models (except the full model where it
industry classifications chosen for each of the
is significant at the 1% level). This is consistent
firms in our sample are too broad to be
with previous research, and also expected. Firms
representative of their actual target markets.
that have been operating longer are more likely
to have finished developing a product and
established distribution channels, leading to Human Resources
higher revenues, less uncertainty and higher firm In model 2 we test the hypotheses developed
valuations compared to younger firms. The about human capital resources and their relation
significance level is though surprising to firm value. Previous research indicates that
considering our small sample size, and to verify human capital resources can lead to sustained
the result we reviewed the coding of the variable competitive advantages (Wright et al., 1994),
to identify possible errors, but none were found. and therefore higher performance and profits.
The authors argue that this can be seen in context Especially inventor presence on the founding
with our model of estimating firm value, where team together with the inventor’s renownedness
as previously mentioned, the value of future and team heterogeneity are highlighted as
growth is not included. Firms that are older have positively related to USO performance (Clarysse
had more time to grow and increase their and Bruneel, 2007; Markman et al., 2008;
revenues, and therefore achieve higher Wright et al., 1994; Hess and Rothaermel, 2012;
valuations in our model, which is reflected in the Zucker et al., 2002; Knockaert et al., 2010). This
high significance level of the control variable led us to respectively develop hypotheses 1a,
age. This however underpins the importance of hypothesis 1b and hypothesis 2.
20 Arif M. Mirza, Ole Rønning

In hypothesis 1a we proposed that USOs that valuable when further developing the
have the technology inventors on the founding technology. Further on, significant technological
team achieve higher valuations in the long term. modifications and developments are likely
In both models 2 and 6, the estimated regression necessary to bring the early-stage technologies
coefficient is positive as proposed, however it is these companies are founded on to commercial
small and insignificant, leading to hypothesis 1a applications, which the technology inventor is
not being supported. The authors propose that likely to be best suited to accomplish. Proxies
the insignificant results might be due to an employed in future research should therefore to a
unfitting operationalization of the inventors better degree measure the actual contribution of
presence in the founding team, and not the fact the inventor, instead of simply her presence.
that the engagement of the technology inventor
In hypothesis 1b we proposed that USOs who
in a USO is not important for the company's
have more renowned inventors should achieve
success, and subsequently firm value. In our
higher firm valuations, and proxied how
sample, almost 90% of the firms had at least one
renowned an inventor is through her citation rate.
of the technology inventors in their founding
Both models 2 and 6 estimate a small,
teams. It is this large percentage rate that makes
insignificant regression coefficient for the effect
it difficult to use the binary operationalization of
of how renowned the inventor is on firm value,
inventor presence to differentiate between USOs
and the coefficient is slightly negative in the full
future firm values, because there simply is not
model. The coefficient itself is highly uncertain
enough variance in the sample. Further on, being
with a large standard error and 95% confidence
named inventor in a business plan and receiving
interval between -0.716 and 0.520. Hypothesis
equity in the new venture does not to a satisfying
1b is followingly not supported. The underlying
degree represent the different levels an inventor
argument for the hypothesis was the signaling
can be engaged in the venture. The effect of the
effect a renowned inventor can have on external
inventors presence on venture performance, and
resource holders. A possible explanation for the
subsequent firm value, is likely to be moderated
small and insignificant regression coefficient is
by other factors such as how much effort and
the conflict of interest between research, and
time the inventor dedicates to the startup. The
commercializing the technology that might arise
technology inventor may solely be given an
for an inventor. Inventors that have high citation
equity stake in the company as a compensation
counts are likely to be avid researchers with a
for the right to use the intellectual property for
focus on publishing their results for the sake of
commercialization purposes. The new venture
research, not commercialization. The shift in
can not, to a high degree, exploit the tacit
mindset needed to go from a research-setting to
knowledge kept by the inventor if the inventor
commercialization-setting is followingly likely
does not really commit herself to the work of
to be more difficult for more renowned investors,
bringing the new product to market. Blair and
which may affect the performance of the new
Hitchens (1998) find that presence of a full-time
venture, and subsequently firm valuation. This is
entrepreneur enhances the performance of USOs
likely to be a larger issue for researchers that
due to signaling effects as well as being able to
have been more active, and published more
focus solely on the new venture. The same
papers. In any case, we do not find support for
arguments can be applied to the technology
including the renownedness of an inventor,
inventor in the founding team. We followingly
proxied through her citation count, as a predictor
conclude that the binary operationalization of
of USO firm value.
inventor participation in the USO founding team
is not suited as a predictor for firm value, and In hypothesis 2 we proposed that founding teams
consequently as a heuristic in a resource-based that are more heterogeneous in terms of
valuation framework, but still argue the educational backgrounds and previous work
importance of inventors in founding teams due to experiences should achieve higher valuations.
the tacit knowledge she holds about the The hypothesis is supported in both model 2 and
underlying technology on which the new venture model 6 at the 10% level. This is in accordance
is founded. Parts of this tacit knowledge is likely with previous research. Visintin and Pittino
to be difficult to make explicit, and for that (2014) find that USOs with founding teams that
reason, the inventor is a non-substitutable asset. have backgrounds from both business and
Not only is the successes of the inventors engineering experience higher sales growths, and
research interesting for the new venture, but the Knockaert et al. (2011) state that these
vast amount of knowledge the inventor has companies receive higher share valuations.
gained through failed experiments is likely to be Having both technical and business knowledge is
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 21

likely to be important in any startup, however, these two variables is natural as founders who
the authors argue that because of the high have worked/researched longer at the university
uncertainty associated with early-stage USOs, are likely to have published more papers and
and because of the challenges they face in terms received more citations. This correlation might
of attracting funding and identifying an have been reduced if another proxy for ‘Ties to
application and target market for the technology, parent’ was used. (Granovetter, 1973) defines the
the combined efforts on both the technological strength of a tie as the combination between the
and business front are likely to be more vital amount of time, the emotional intensity, the
than in the average startup. For example, the intimacy, and the reciprocal services that
long development times (Rasmussen et al., 2013) characterize each tie, and by only using the time
are likely to require a greater degree of aspect of this definition, we might not represent
synchronization between the development plan the strength of the tie adequately. A second
and financial plan to assure enough funding reason for the loss of significance may be due to
throughout the development process. The the small sample size. As (Field, 2013) points
general-purpose nature of the technologies out, small or medium sized effects will not be
(Nelsen, 1991; del Campo et al., 1999) require a significant in our model.
larger degree of cooperation between business
Based on the above discussion, we can not
people and technical people to assure that the
advocate that the strength of ties between a USO
technology is developed towards applications
and the parent university, proxied through the
that provide the largest benefits to the customer
number of years the founders have worked/done
as well as the largest markets. Further on, the
research at the parent, is suited as a heuristic in a
early-stage nature of the technologies (Shane,
resource-based valuation framework, but we do
2004) to a greater degree are likely to require
attain our belief that this specific resource can
these companies to attract partners, be that
positively affect the performance of an USO, and
financial or industrial, and business acumen in
encourage the search of a more fitting proxy for
the top management team is therefore as
the relationship. Seen from a resource-based
important as technical knowledge. In conclusion,
perspective, a strong affiliation between the USO
our results indicate heterogeneity in the founding
and its parent university can be valuable because
team of a USO as a valid predictor for firm
the parent can provide access to resources such
value, and that it therefore should be included as
as talent and critical financial resources
a heuristic in a resource-based valuation
(Bigliardi et al., 2013), knowledge and
framework.
complementary R&D (Lubik et al., 2013), labs
and expensive equipment at low cost (Starr and
Social Resources MacMillan, 1990; Roberts and Malonet, 1996).
Model 3 tests hypotheses developed about the Further on, the university has a limited set of
relationship between initial social resource- resources, and the authors argue that the stronger
endowments in a USO and firm value. Social the social ties between the founders and the
resources refer to the personal networks of the parent, the more likely the USO is to receive a
founding team, and in the performance literature larger share of these resources because the
on USOs, the degree of university affiliation has founders can draw upon their network. The
been highlighted as correlated with performance, affiliation is likely to be rare due to the
as has the quality of the university itself. university only supporting USOs that originate
within its own walls. Inimitability is likely due to
In hypothesis 3a we proposed that USOs with the relationship between the university and the
stronger ties to the parent university in their early new venture being built over time and being path
stages should be valued higher than those with dependent (Zander and Zander, 2005; Santala
weaker ties. The strength of the tie is proxied and Parvinen, 2007), i.e. the relationship is built
through the total number of years the founders through the work the technology inventor
have worked/researched at the parent university. conducted at the university and through the
The hypothesis is supported in model 3 at the 1% inventors interactions with other people at the
level, but looses its significance in the full university during her time there. Finally, the
model. The loss of significance can be attributed relationship between the parent university and
to the correlation between ‘Ties to parent’ and the USO is non-substitutable due to each
‘Inventor renownedness’, and the variable does university’s uniqueness in terms of quality and
indeed keep its significance when the full model expertise. Future research should therefore seek
is estimated without ‘Inventor renownedness’ to develop better proxies to the strength of social
being included. The high correlation between ties between a USO and its parent to test if this
22 Arif M. Mirza, Ole Rønning

variable is suited as a heuristic in a resource- acquire other resources that can be VRIN. In
based valuation framework. agreement with this, our results indicate that the
binary operationalization of received investor
In hypothesis 3b we proposed that USOs
capital in the early-stages of a USOs life-cycle is
originating from more renowned, higher quality
not a good predictor for the future value of the
universities, should achieve higher valuations,
firm, which has important implications. Previous
and proxied the quality of the university through
research has focused on the relationship between
its world rank. The hypothesis is supported in
investor capital and performance, and has found
both model 3 and the full model at the 10% and
that there is a positive significant relationship
5% level respectively. Existing streams of
between the two. However, performance has
research highlight access to expert knowledge,
commonly been measured through sales growth
people (Powers and McDougall, 2005), and
(e.g. Walter et al. (2006), Lubik et al. (2013) and
funding (Di Gregorio and Shane, 2003) as three
Bigliardi et al. (2013)), firm outcome, i.e. if the
reasons for why the quality of the university
firm achieved IPO, got acquired, survived or
affects USO performance. The authors further
failed (e.g. Criaco et al. (2013) and Fuller and
argue that higher quality universities are likely to
Rothaermel (2012)), or number of employees
produce higher quality research, and followingly
and employment growth (e.g. Sternberg (2014),
better technology on which USOs can be
Bigliardi et al. (2013), Zerbinati et al. (2012),
founded. Because most of these companies are
Zhang (2009) and Walter et al. (2006)). Some of
solely founded around a technology, the quality
these measures have obvious positive
of the technology is obviously important to firm
relationships to a firm attaining investor capital.
performance, and subsequently firm value.
For example, investor capital enables a company
Higher quality universities are also likely to have
to hire more employees which obviously lead to
stronger signaling effects towards external
employee growth, it may boost sales through
resource holders. These signaling effects can be
marketing, or it may keep the company operating
particularly valuable for early stage USOs
longer despite large losses. Our study, which
because of no objective operating data, and
instead of indirect performance measures relates
consequently investors not being able to
‘Investor capital’ in an early-stage to actual firm
adequately assess the technology on which the
value, shows that even though capital may
new venture is founded (Wright et al., 2006).
increase firm performance on certain measures
Our analysis followingly indicates that the
used by other researchers, it does not necessarily
quality of the university should indeed be
increase firm value. In fact, we find a negative
included as a heuristic in a resource-based
relationship between the two (although the
valuation framework.
relationship is insignificant).
Financial Resources There are multiple reasons for why early-stage
investments may negatively impact the
In model 4 we test hypothesis 4, where we performance, and subsequently value, of USOs.
propose that USOs who receive investor capital First of all, investors are generally interested in
in their early-stages, proxied as a binary variable, the largest return possible and, considering the
should achieve higher valuations than those that time value of money, at the earliest possible time
do not. Financial resources are for obvious after investing. Early-stage USOs on the other
reasons important for all early stage ventures, hand have long development times (Rasmussen
however, as highlighted in the literature review, et al., 2013; Lawton Smith and Ho, 2006). This
it is especially critical for early stage USOs due interest gap may lead to pressure from the
to the large development costs they face. investors on the company to speed up
Surprisingly the regression coefficient is development, take shortcuts where possible,
negative and insignificant in both model 4 and make rash decisions, and followingly harm the
the full model, and we do not find support for new venture's performance and value. Another
hypothesis 4. possible reason for the negative relationship is
Although surprising, the result is consistent with the competence and knowledge gap between
the previous discussion on financial resources investors and founders in USOs. In research on
not on their own being valuable, rare, inimitable the effect of venture capital it has been
and non-substitutable (VRIN), and should highlighted that venture capitalists contribute
therefore not lead to sustained competitive with more than simply money. Their networks,
advantages, and followingly higher profits and experience and advice are almost equally, if not
higher value. Instead, capital enables firms to more, important in helping the new venture
succeed (Hellman and Puri, 2000; Sapienza,
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 23

1992). Many USOs on the other hand are and inimitable (Markman et al., 2004). As for the
working with complex technology that investors patents non-substitutability, Markman et al.
may have problems understanding (Shane, (2004) operationalize non-substitutability as the
2004). This can make it difficult for the investors patents number of claims, with the rationale
to contribute with this added value, and may being that claims define the scope of an
ultimately lead them to give bad advice. invention and distinguish its property from the
surrounding technological territory, and once a
On the other hand, there can be simple
technology space is protected by a patent,
methodological explanations to our results. Our
substitution becomes difficult and costly. Our
binary operationalization is crude, and does not
analysis confirms the positive relationship
incorporate the actual amount received, or who
between patents and the long term value of a
the investor is. These two variables are likely to
USO, and establishes a theoretical linkage
have strong moderating effects on the investment
between the two. We conclude that the number
variable. Although our results are surprising, the
of patents filed or published by a USO during its
authors find them plausible. The analysis
first two years should be included as a heuristic
indicates that investor capital, proxied through a
in a resource-based valuation framework.
binary variable, is not a good predictor for early-
stage USO firm value, and therefore should not
be included as a heuristic in a resource-based Conclusion and Limitations
valuation framework.
The topic of valuation has traditionally been a
field within corporate finance, with common
Technological Resources valuation methods being dependent on future
The last category of technological resources are cash flows, assets or valuations of comparative
particularly important for early stage USOs, as firms (Seppä and Laamanen, 2001). However, as
they are founded to commercialize distinct Mirza and Rønning (2015) point out, these
technological innovations developed at the methods are poorly suited to be used on early-
parent university (O’Shea et al., 2008). In model stage USOs due to a fundamental mismatch
5 we test hypothesis 5, where we propose that between the required inputs and the operating
USOs who publish or file more patents during data available for these companies. This study
their early stages achieve higher valuations. The has taken a resource-based perspective on
hypothesis is supported in both model 5 and the valuation to develop a conceptual framework
full model at the 5% level. where resources are used as valuation inputs
rather than financial data. The underlying
The authors argue that patents are positively argument being that when it is difficult to value a
related to USO value because they are a resource company based on outputs (cash flows etc.), a
that, in accordance with Mirza and Rønning method based on inputs (resources, founder
(2015), can be valuable, rare, inimitable and non- characteristics, network etc.) that can be
substitutable, and followingly result in a objectively measured, and applied as heuristics,
sustained competitive advantage, at least for the may prove more satisfactory. To empirically test
duration of the patents validity. The reviewed the framework we have leveraged existing
literature has shown that the patents of the new research on the resource-performance
venture are likely to be valuable (Deeds et al., relationships in USOs. This led us to develop six
2000; Zahra and Bogner, 2000; Powers and hypotheses about the relationship between
McDougall, 2005). The patents protect the new human capital, social, financial and
venture from competitors, as well as it is natural technological resources and firm value. The
to believe that the new venture would not have hypotheses were tested using multiple
been founded if the founders did not believe that hierarchical regression on a data sample of 63
the underlying technology, protected by the Norwegian USOs from the EntPro project
patents, would yield a positive and significant founded in the period 2000 - 2005. Our analysis
financial return. Further on, the larger the scope has shown that there is a significant positive
and number of patents, the greater is the relationship between the long term equity value
likelihood of protecting several application areas, of a USO and the founding teams heterogeneity,
which in the future can lead to licensing incomes the quality of the parent university and the
from others who wish to utilize the patents, or a number of patents the new venture has filed or
protected technology in multiple markets, which published. We did not find a significant
naturally is valuable. For obvious reasons, and relationship between firm value and inventors on
per definition, the patents are likely to be rare the founding team, USOs with stronger ties to
24 Arif M. Mirza, Ole Rønning

Figure 7. Conceptual framework

the parent, or USOs who attain early-stage confident they have, but to develop a complete
investor capital. For the latter three hypotheses valuation framework based on resources will
that were not supported we have argued that this require significant attention from researchers in
might be due to unrepresentative both strategic management and corporate finance
operationalizations, and further research is to adequately bridge the gap between the two.
therefore necessary to verify our results. For the
This study has a number of implications for both
former three hypotheses, the authors argue that
investors, entrepreneurs and researchers. For
given the significant relationship identified,
investors, we have established the initial
USOs who have more heterogeneous founding
theoretical linkages between initial resource
teams, originate from higher quality universities
endowments and subsequent firm value, which
and have more filed or published patents in their
can be used as heuristics when valuing an early-
early stages, should, ceteris paribus, be valued
stage USO. Instead of founders and investors
higher than those that do not.
estimating and forecasting cash flows that are
This is because, as resource-based theory impossible to reliably estimate in the early stages
suggests, resources are what ultimately drives (Mirza and Rønning, 2015), and trying to justify
value creation in firms, and our analysis uncertain growth- and discount rates where a
indicates that these specific resources are small change in either can result in large changes
associated with higher long term firm values. in the firm value, using resources which both the
Given the uncertain nature of these firms, it investor and the entrepreneurs can objectively
makes more sense to use these resources to guide measure and agree upon, is likely to be a more
valuations, and distinguish between different effective and fair process. Although we have not
investment opportunities, instead of trying to developed a complete valuation framework,
estimate absolute firm values with uncertain cash investors can use the conceptual framework to
flows and discount rates. On a conceptual level, take into consideration the identified
this is highlighted in the framework in figure 7, relationships when estimating firm values.
and substantiated, although tentative, through Further on, we have shown that resource-
our empirical analysis. As such, these resource performance relationships highlighted in theory
characteristics should be applicable to be used as are not necessarily good heuristics for predicting
heuristics to value a USO in its early stages the future value of a firm, such as the presence of
regardless of the high fundamental uncertainty an investor in the founding team or early-stage
surrounding these firms. This is however the first investor capital. Rather it is necessary to dig
study of its kind, and like traditional valuation deeper into the resources to assert their effect on
methods weren't developed in a single study, firm performance and subsequent firm value. For
neither will a resource-based valuation entrepreneurs, the findings highlight initial
framework. We set out to prove the feasibility of resources that the new venture should strive to
such a framework, which the authors feel assemble to increase their performance, and
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 25

subsequently the value of their firm. We have venture capitalists. Further on, data regarding
shown that entrepreneurs should struggle to founder backgrounds, financing and patents
assemble complementary and heterogeneous should be collected first hand directly from the
founding teams to achieve higher firm USO to avoid information errors.
valuations, and that asserting control over their
This study, as the first of its kind, aimed to
technology through patents is positively related
establish the initial theoretical linkages between
to firm value. Moreover, entrepreneurs licensing
resources and firm value, and as such start to
technology from universities should take into
bridge the gap between traditional corporate
account the quality of the university when
finance, strategic management and USOs. There
deciding on what technology to license. Finally,
is significant research necessary to develop a
our results pave the way for an entire new stream
complete standardized valuation framework with
of research bridging the gap between strategic
a satisfying degree of predictive power. It is
management and traditional corporate finance.
however clear that today’s valuation methods are
not good enough, and that a resource-based
Further Research valuation framework has potential, so research
efforts toward this should be prioritized.
The limitations previously highlighted limit the
generalizability of this study and underpin the
need to verify the results obtained using larger References
samples. Future research on the topic should
AGARWAL, R. & SHAH, S. K. 2014. Knowledge
however not only increase the sample size, in sources of entrepreneurship: Firm
absolute numbers as well as in geographical formation by academic, user and
spread, but also the number of resource-value employee innovators. Research Policy.
relationships examined. We have only been able ALVAREZ, S. A. & BUSENITZ, L. W. 2001. The
to examine six of these relationships. To develop entrepreneurship of resource-based theory.
a complete resource-based valuation framework Journal of management, 27, 755-775.
it will be necessary to significantly expand on BAKER, M. & RUBACK, R. 1999. Estimating
this, find better and more detailed proxies which industry multiples. Harvard University,
to a greater degree can differentiate between Google Scholar.
USOs, as well as examine the relative BARNEY, J. 1991. Firm resources and sustained
importance of each resource type. It is also competitive advantage. Journal of
necessary to investigate the moderating effect of management, 17, 99-120.
firm capabilities on the relationship between BARNEY, J., WRIGHT, M. & KETCHEN, D. J.
resources and firm value, and whether it is 2001. The resource-based view of the
possible to objectively operationalize these firm: Ten years after 1991. Journal of
capabilities to be included in a resource-based management, 27, 625-641.
valuation framework. Further on, we have not BARNEY, J. B. & CLARK, D. N. 2007. Resource-
incorporated the possible effects of external based theory: Creating and sustaining
competitive advantage, Oxford University
factors, as highlighted by the Industrial
Press Oxford.
organization perspective, on the value of a firm
BIDAULT, F. & CUMMINGS, T. 1994.
in our framework. Neither have we included
Innovating through alliances: expectations
variables such as estimated development times and limitations. R&D Management, 24,
and capital needs which are likely to have strong 033-045.
implications on value estimations. To develop a BIERLY, P. E., DAMANPOUR, F. & SANTORO,
complete and robust framework, such variables M. D. 2009. The application of external
and their effects on early stage value must be knowledge: organizational conditions for
investigated. exploration and exploitation. Journal of
Amongst our results, although statistically Management Studies, 46, 481-509.
BIGLIARDI, B., GALATI, F. & VERBANO, C.
insignificant, we find the negative relationship
2013. Evaluating performance of
between early stage investor capital and long
university spin-off companies: Lessons
term value especially interesting. This finding from italy. Journal of technology
contradicts existing research on the subject, and management & innovation, 8, 178-188.
researchers should try to verify this, as the BLAIR, D. M. & HITCHENS, D. M. 1998.
implications of this result for entrepreneurs are Campus Companies--UK and Ireland,
significant. It is also important to verify our Ashgate Aldershot, UK.
results using different data sources. First of all,
valuations should be conducted by impartial
26 Arif M. Mirza, Ole Rønning

BONARDO, D., PALEARI, S. & VISMARA, S. entrepreneurship. The Journal of

2011. Valuing University Based Firms: Technology Transfer, 35, 1-15.
The Effects of Academic Affiliation on COLOMBO, M. G. & GRILLI, L. 2010. On
IPO Performance. Entrepreneurship growth drivers of high-tech start-ups:
Theory and Practice, 35, 755-776. Exploring the role of founders' human
BORLAUG, S. B. & STEP, N. 2009. Between capital and venture capital. Journal of
entrepreneurship and technology transfer: Business Venturing, 25, 610-626.
Evaluation of the FORNY programme, COLOMBO, M. G. & PIVA, E. 2012. Firms’
NIFU STEP. genetic characteristics and competence-
BOUDREAU, J. W. 1983. Economic enlarging strategies: A comparison
considerations in estimating the utility of between academic and non-academic
human resource productivity improvement high-tech start-ups. Research Policy, 41,
programs. Personnel Psychology, 36, 551- 79-92.
576. COOK, R. D. & WEISBERG, S. 1982. Residuals
BOUDREAU, J. W. & BERGER, C. J. 1985. and influence in regression.
Decision-theoretic utility analysis applied CRESWELL, J. W. 2003. Research design.
to employee separations and acquisitions. Qualitative, quantitative and mixed
Journal of Applied Psychology, 70, 581. methods approaches.
BOWERMAN, B. L. & O'CONNELL, R. T. 1990. CRIACO, G., MINOLA, T., MIGLIORINI, P. &
Linear statistical models: An applied SERAROLS-TARRÉS, C. 2013. “To have
approach, PWS-Kent Boston. and have not”: founders’ human capital
BRUSH, C. G., GREENE, P. G. & HART, M. M. and university start-up survival. The
2001. From initial idea to unique Journal of Technology Transfer, 1-27.
advantage: The entrepreneurial challenge DAMODARAN, A. 2005. Marketability and value:
of constructing a resource base. The Measuring the illiquidity discount.
Academy of Management Executive, 15, Available at SSRN 841484.
64-78. DAMODARAN, A. 2009. Valuing young, start-up
BRÜDERL, J., PREISENDÖRFER, P. & and growth companies: estimation issues
ZIEGLER, R. 1992. Survival chances of and valuation challenges. Stern School of
newly founded business organizations. Business, New York University (New
American sociological review, 227-242. York).
CARPENTER, M. A. & FREDRICKSON, J. W. DAMODARAN, A. 2012. Investment valuation:
2001. Top management teams, global Tools and techniques for determining the
strategic posture, and the moderating role value of any asset, John Wiley & Sons.
of uncertainty. Academy of Management DE CLERCQ, D., FRIED, V. H., LEHTONEN, O.
journal, 44, 533-545. & SAPIENZA, H. J. 2006. An
CLARYSSE, B. & BRUNEEL, J. 2007. Nurturing entrepreneur's guide to the venture capital
and growing innovative start ups: the galaxy. The Academy of Management
role of policy as integrator. R&d Perspectives, 20, 90-112.
Management, 37, 139-149. DEEDS, D. L., DECAROLIS, D. & COOMBS, J.
CLARYSSE, B. & MORAY, N. 2004. A process 2000. Dynamic capabilities and new
study of entrepreneurial team formation: product development in high technology
the case of a research-based spin-off. ventures: an empirical analysis of new
Journal of Business Venturing, 19, 55-79. biotechnology firms. Journal of Business
CLARYSSE, B., WRIGHT, M., LOCKETT, A., venturing, 15, 211-229.
MUSTAR, P. & KNOCKAERT, M. 2007. DEL CAMPO, A. A., SPARKS, A., HILL, R. C. &
Academic spin-offs, formal technology KELLER, R. T. 1999. The transfer and
transfer and capital raising. Industrial and commercialization of university-
Corporate Change, 16, 609-640. developed medical imaging technology:
CLARYSSE, B., WRIGHT, M. & VAN DE opportunities and problems. Engineering
VELDE, E. 2011. Entrepreneurial Origin, Management, IEEE Transactions on, 46,
Technological Knowledge, and the 289-298.
Growth of Spin Off Companies. Journal DENG, M., EASTON, P. D. & YEO, J. 2012.
of Management Studies, 48, 1420-1442. Another look at equity and enterprise
COLE, S. & COLE, J. R. 1967. Scientific output valuation based on multiples. Available at
and recognition: A study in the operation SSRN 1462794.
of the reward system in science. American DI GREGORIO, D. & SHANE, S. 2003. Why do
sociological review, 377-390. some universities generate more start-ups
COLOMBO, M., MUSTAR, P. & WRIGHT, M. than others? Research policy, 32, 209-227.
2010. Dynamics of science-based
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 27

DIAMOND, A. M. 1984. An economic model of moves. Administrative science quarterly,

the life-cycle research productivity of 659-684.
scientists. Scientometrics, 6, 189-196. HAYTER, C. S. 2013. Harnessing University
DJOKOVIC, D. & SOUITARIS, V. 2008. Spinouts Entrepreneurship for Economic Growth
from academic institutions: a literature Factors of Success Among University
review with suggestions for further Spin-offs. Economic Development
research. The Journal of Technology Quarterly, 27, 18-28.
Transfer, 33, 225-247. HEIRMAN, A. & CLARYSSE, B. 2004. How and
DOUTRIAUX, J. & BARKER, M. 1995. The why do research-based start-ups differ at
university-industry relationship in science founding? A resource-based
and technology. Gouvernement du configurational perspective. The Journal
Canada-Industry Canada. of Technology Transfer, 29, 247-268.
DURBIN, J. & WATSON, G. S. 1951. Testing for HELLMAN, T. & PURI, M. 2000. The interaction
serial correlation in least squares between product market and financing
regression. II. Biometrika, 159-177. strategy: the role of venture capital.
FIELD, A. 2013. Discovering statistics using IBM Review of Financial Studies, 13, 959-984.
SPSS statistics, Sage. HESS, A. & ROTHAERMEL, F. T. 2012.
FLORIN, J., LUBATKIN, M. & SCHULZE, W. Intellectual human capital and the
2003. A social capital model of high- emergence of biotechnology: trends and
growth ventures. Academy of Management patterns, 1974–2006. Engineering
Journal, 46, 374-384. Management, IEEE Transactions on, 59,
FREEMAN, J. 1999. Venture capital as an 65-76.
economy of time. Corporate social capital HIGGINS, M. J., STEPHAN, P. E. & THURSBY,
and liability. Springer. J. G. 2011. Conveying quality and value in
FULLER, A. W. & ROTHAERMEL, F. T. 2012. emerging industries: Star scientists and the
When stars shine: the effects of faculty role of signals in biotechnology. Research
founders on new technology ventures. Policy, 40, 605-617.
Strategic Entrepreneurship Journal, 6, HOANG, H. & ANTONCIC, B. 2003. Network-
220-235. based research in entrepreneurship: A
GE, D., MAHONEY, J. M. & MAHONEY, J. T. critical review. Journal of business
2005. New Venture Valuation by Venture venturing, 18, 165-187.
Capitalists: An Integrative Approach. HSU, D. H. & ZIEDONIS, R. H. PATENTS AS
University of Illinois at Urbana- QUALITY SIGNALS FOR
Champaign, College of Business. ENTREPRENEURIAL VENTURES.
GIGERENZER, G. & GAISSMAIER, W. 2011. Academy of Management Proceedings,
Heuristic decision making. Annual review 2008. Academy of Management, 1-6.
of psychology, 62, 451-482. JENSEN, A. R. 1980. Bias in mental testing.
GIGERENZER, G., HERTWIG, R. & PACHUR, JR, A. M. D. 1986. What is a Citation Worth? The
T. 2011. Heuristics: The Foundations of Journal of Human Resources, 21, 200-
Adaptive Behavior, Oxford University 215.
Press, Inc. KAMIYAMA, S. & SHEEHAN, J. 2006.
GOMPERS, P. A. & LERNER, J. 1999. What Valuation And Exploitation of Intellectual
drives venture capital fundraising? : Property. Statistical Analysis of Science,
National bureau of economic research. Technology and Industry. Working Paper
GRANDI, A. & GRIMALDI, R. 2005. Academics' 2006/5. Paris, Organisation for Economic
organizational characteristics and the Co-operation and Development (OECD).
generation of successful business ideas. KATZ, R. L. 1974. Skills of an effective
Journal of Business Venturing, 20, 821- administrator, Harvard Business Press.
845. KEELEY, R. H., PUNJABI, S. & TURKI, L. 1996.
GRANOVETTER, M. S. 1973. The strength of Valuation of early-stage ventures: option
weak ties. American journal of sociology, valuation models vs. traditional
1360-1380. approaches. The Journal of
GURDON, M. A. & SAMSOM, K. J. 2010. A Entrepreneurial Finance, 5, 115-138.
longitudinal study of success and failure KNOCKAERT, M., CLARYSSE, B. & WRIGHT,
among scientist-started ventures. M. 2010. The extent and nature of
Technovation, 30, 207-214. heterogeneity of venture capital selection
HAMBRICK, D. C., CHO, T. S. & CHEN, M.-J. behaviour in new technology based
1996. The influence of top management firms. R&d Management, 40, 357-371.
team heterogeneity on firms' competitive KNOCKAERT, M., LOCKETT, A., CLARYSSE,
B. & WRIGHT, M. 2006. Do human
28 Arif M. Mirza, Ole Rønning

capital and fund characteristics drive capitalists: an empirical study. Venture

follow-up behaviour of early stage high- Capital, 14, 151-174.
tech VCs? International Journal of MIRZA, A. M. & RØNNING, O. 2015. Examining
Technology Management, 34, 7-27. Equity Valuation of University Spin-off
KNOCKAERT, M., UCBASARAN, D., WRIGHT, Companies: A Literature Review.
M. & CLARYSSE, B. 2011. The MULLEN, M. R., BUDEVA, D. G. & DONEY, P.
Relationship Between Knowledge M. 2009. Research Methods in the
Transfer, Top Management Team Leading Small Business–Entrepreneurship
Composition, and Performance: The Case Journals: A Critical Review with
of Science-Based Entrepreneurial Firms. Recommendations for Future Research*.
Entrepreneurship Theory and Practice, Journal of Small Business Management,
35, 777-803. 47, 287-307.
KOR, Y. Y. & MAHONEY, J. T. 2004. Edith MUSTAR, P., RENAULT, M., COLOMBO, M.
Penrose's (1959) Contributions to the G., PIVA, E., FONTES, M., LOCKETT,
Resource-based View of Strategic A., WRIGHT, M., CLARYSSE, B. &
Management. Journal of Management MORAY, N. 2006. Conceptualising the
Studies, 41, 183-191. heterogeneity of research-based spin-offs:
KOTTER, J. P. 1982. What effective general A multi-dimensional taxonomy. Research
managers really do, Harvard Business policy, 35, 289-308.
Review. NARAYANAN, V. K. 2000. Managing technology
LAWTON SMITH, H. & HO, K. 2006. Measuring and innovation for competitive advantage.
the performance of Oxford University, NECK, H. M., MEYER, G. D., COHEN, B. &
Oxford Brookes University and the CORBETT, A. C. 2004. An
government laboratories’ spin-off entrepreneurial system view of new
companies. Research Policy, 35, 1554- venture creation. Journal of Small
1568. Business Management, 42, 190-208.
LIU, J., NISSIM, D. & THOMAS, J. 2002. Equity NELSEN, L. L. 1991. The lifeblood of
valuation using multiples. Journal of biotechnology: University-industry
Accounting Research, 40, 135-172. technology transfer. The business of
LONG, C. 2002. Patent signals. The University of biotechnology: from the bench to the
Chicago Law Review, 625-679. street, 39-75.
LUBIK, S., GARNSEY, E., MINSHALL, T. & NELSON, R. E. 1989. The Strength of Strong Ties:
PLATTS, K. 2013. Value creation from Social Networks and Intergroup Conflict
the innovation environment: partnership in Organizations. Academy of
strategies in university spin outs. R&D Management Journal, 32, 377-401.
Management, 43, 136-150. NEWBERT, S. L. 2007. Empirical research on the
MAHONEY, J. T. & PANDIAN, J. R. 1992. The resource based view of the firm: an
resource based view within the assessment and suggestions for future
conversation of strategic management. research. Strategic management journal,
Strategic management journal, 13, 363- 28, 121-146.
380. O'SHEA, R. P., ALLEN, T. J., CHEVALIER, A. &
MARKMAN, G. D., ESPINA, M. I. & PHAN, P. ROCHE, F. 2005. Entrepreneurial
H. 2004. Patents as Surrogates for orientation, technology transfer and
Inimitable and Non-Substitutable spinoff performance of US universities.
Resources. Journal of Management, 30, Research Policy, 34, 994-1009.
529-544. O’SHEA, R. P., CHUGH, H. & ALLEN, T. J.
MARKMAN, G. D., SIEGEL, D. S. & WRIGHT, 2008. Determinants and consequences of
M. 2008. Research and technology university spinoff activity: a conceptual
commercialization. Journal of framework. The Journal of Technology
Management Studies, 45, 1401-1423. Transfer, 33, 653-666.
MERGES, R. P. & NELSON, R. R. 1990. On the OLOFSSON, C. & WAHLBIN, C. 1984.
complex economics of patent scope. Technology-based new ventures from
Columbia Law Review, 839-916. technical universities: A Swedish case.
MILLER, D. 2003. An asymmetry based view of Frontiers of entrepreneurship research,
advantage: towards an attainable 192-211.
sustainability. Strategic management PENROSE, E. T. 1959. The Theory of the Growth
journal, 24, 961-976. of theFirm. London:
MILOUD, T., ASPELUND, A. & CABROL, M. BlackwellsPenroseThe Theory of Growth
2012. Startup valuation by venture of the Firm1959.
 Resource-based Equity Valuation of University Spin-off Companies: An empirical study 29

PENROSE, E. T. 1995. The Theory of the Growth SCHERER, F. M. & ROSS, D. 1990. Industrial
of the Firm, Oxford university press. market structure and economic
PÉREZ PÉREZ, M. & SÁNCHEZ, A. M. 2003. performance. University of Illinois at
The development of university spin-offs: Urbana-Champaign's Academy for
early dynamics of technology transfer and Entrepreneurial Leadership Historical
networking. Technovation, 23, 823-831. Research Reference in Entrepreneurship.
POLANYI, M. 2012. Personal knowledge: SEPPÄ, T. J. & LAAMANEN, T. 2001. Valuation
Towards a post-critical philosophy, of venture capital investments: empirical
University of Chicago Press. evidence. R&D Management, 31, 215-
PORTER, M. E. 1981. The contributions of 230.
industrial organization to strategic SHANE, S. 2001. Technology regimes and new
management. Academy of management firm formation. Management science, 47,
review, 6, 609-620. 1173-1190.
POWERS, J. B. & MCDOUGALL, P. P. 2005. SHANE, S. 2004. Academic entrepreneurship:
University start-up formation and University spinoffs and wealth creation,
technology licensing with firms that go Edward Elgar Publishing.
public: a resource-based view of academic SHANE, S. & CABLE, D. 2002. Network Ties,
entrepreneurship. Journal of Business Reputation, and the Financing of New
Venturing, 20, 291-311. Ventures. Management Science, 48, 364-
RASMUSSEN, E., BORLAUG, S. B. & 381.
BULANOVA, O. 2013. Verdiskaping i SHANE, S. & STUART, T. 2002. Organizational
forskningsbaserte selskaper og lisenser Endowments and the Performance of
støttet av FORNY-programmet. University Start-ups. Management
RASMUSSEN, E., BULANOVA, O., JENSEN, A. Science, 48, 154-170.
& CLAUSEN, T. 2012. The Impact of SHORT, J. C., KETCHEN, D. J., COMBS, J. G. &
Science-Based Entrepreneurial Firms-a IRELAND, R. D. 2010. Research methods
Literature Review and Policy Synthesis. in entrepreneurship Opportunities and
Report 3-2012, 154. challenges. Organizational Research
ROBERTS, E. B. 1991. Entrepreneurs in high Methods, 13, 6-15.
technology: Lessons from MIT and SILBER, W. L. 1991. Discounts on restricted
beyond, Oxford University Press New stock: The impact of illiquidity on stock
York. prices. Financial Analysts Journal, 47, 60-
ROBERTS, E. B. & HAUPTMAN, O. 1986. The 64.
process of technology transfer to the new SPENCE, A. M. & MICHAEL, A. 1974. Market
biomedical and pharmaceutical firm. signaling: Informational transfer in hiring
Research Policy, 15, 107-119. and related screening processes, Harvard
ROBERTS, E. B. & MALONET, D. E. 1996. University Press Cambridge, MA.
Policies and structures for spinning off STARR, J. A. & MACMILLAN, I. 1990. Resource
new companies from research and cooptation via social contracting:
development organizations#. R&D Resource acquisition strategies for new
Management, 26, 17-48. ventures. Strategic Management Journal,
ROTHAERMEL, F. T., AGUNG, S. D. & JIANG, 11, 79-92.
L. 2007. University entrepreneurship: a STEFFY, B. D. & MAURER, S. D. 1988.
taxonomy of the literature. Industrial and Conceptualizing and measuring the
corporate change, 16, 691-791. economic effectiveness of human resource
ROTHAERMEL, F. T. & THURSBY, M. 2005. activities. Academy of Management
Incubator firm failure or graduation?: The Review, 13, 271-286.
role of university linkages. Research STERNBERG, R. 2014. Success factors of
policy, 34, 1076-1090. university-spin-offs: Regional government
ROURE, J. B. & KEELEY, R. H. 1990. Predictors support programs versus regional
of success in new technology based environment. Technovation, 34, 137-148.
ventures. Journal of business venturing, 5, STEVENS, J. P. 2012. Applied multivariate
201-220. statistics for the social sciences,
SANTALA, M. & PARVINEN, P. 2007. From Routledge.
strategic fit to customer fit. Management STUART, T. E. 2000. Interorganizational alliances
Decision, 45, 582-601. and the performance of firms: A study of
SAPIENZA, H. J. 1992. When do venture growth and innovation rates in a high-
capitalists add value? Journal of Business technology industry. Strategic
Venturing, 7, 9-27. management journal, 21, 791-811.
30 Arif M. Mirza, Ole Rønning

STUDENMUND, A. H. 2011. Using econometrics: moderating effect of the competitive

A practical guide. environment. Journal of business
SZILAGYI, A. D., JR. & SCHWEIGER, D. M. venturing, 15, 135-173.
1984. Matching Managers to Strategies: A ZANDER, I. & ZANDER, U. 2005. The Inside
Review and Suggested Framework. The Track: On the Important (But Neglected)
Academy of Management Review, 9, 626- Role of Customers in the Resource
637. Based View of Strategy and Firm
TABACHNICK, B. & FIDELL, L. 1996. Using Growth*. Journal of Management Studies,
multivariate statistics. 42, 1519-1548.
TIMMONS, J. A. & SPINELLI, S. 2004. New ZERBINATI, S., SOUITARIS, V. & MORAY, N.
venture creation: entrepreneurship for the 2012. Nurture or nature? The growth
21st century, McGraw-Hill/Irwin. paradox of research-based spin-offs.
VISINTIN, F. & PITTINO, D. 2014. Founding Technology Analysis & Strategic
team composition and early performance Management, 24, 21-35.
of university—Based spin-off companies. ZHANG, J. 2009. The performance of university
Technovation, 34, 31-43. spin-offs: an exploratory analysis using
VOHORA, A., LOCKETT, A. & WRIGHT, M. venture capital data. The Journal of
Critical junctures in the growth in Technology Transfer, 34, 255-285.
university high-tech spinout companies. ZUCKER, L. G., DARBY, M. R. &
Conference paper at Frontiers of ARMSTRONG, J. 1998. Geographically
Entrepreneurship Research, University of localized knowledge: spillovers or
Colorado, 2002. markets? Economic Inquiry, 36, 65-86.
WALTER, A., AUER, M. & RITTER, T. 2006. ZUCKER, L. G., DARBY, M. R. &
The impact of network capabilities and ARMSTRONG, J. S. 2002.
entrepreneurial orientation on university Commercializing knowledge: University
spin-off performance. Journal of Business science, knowledge capture, and firm
Venturing, 21, 541-567. performance in biotechnology.
WHITLEY, R. 1989. On the nature of managerial Management Science, 48, 138-153.
tasks and skills: their distinguishing
characteristics and organization. Journal
of Management Studies, 26, 209-224.
WIDDING, L. O., MATHISEN, M. T. &
MADSEN, O. 2009. University-affiliated
venture capital funds: Funding of
university spin-off companies.
International Journal of Technology
Transfer and Commercialisation, 8, 229-
245.
WILKINS, A. L. 1989. Developing corporate
character: How to successfully change an
organization without destroying it, Jossey-
Bass.
WRIGHT, M., CLARYSSE, B. & MOSEY, S.
2012. Strategic entrepreneurship, resource
orchestration and growing spin-offs from
universities. Technology Analysis &
Strategic Management, 24, 911-927.
WRIGHT, M., LOCKETT, A., CLARYSSE, B. &
BINKS, M. 2006. University spin-out
companies and venture capital. Research
Policy, 35, 481-501.
WRIGHT, P. M., MCMAHAN, G. C. &
MCWILLIAMS, A. 1994. Human
resources and sustained competitive
advantage: a resource-based perspective.
International journal of human resource
management, 5, 301-326.
ZAHRA, S. A. & BOGNER, W. C. 2000.
Technology strategy and software new
ventures' performance: exploring the
!
!
Appendix
!

Article 1
Appendix 1 – Search Phrase………………………………….1

Appendix 2 – Articles From Structured Literature Search…...2

Article 2
Appendix 3 - Codebook………………………………………5

Appendix 4 – Descriptives……………………………………8

Appendix 5 - SPSS Model Information……..………………..9

Appendix 6 - Normality of residuals………………………...16

Appendix 7: Illiquidity Discount…………………………….17

1! !

Appendix 1 - Search Phrase

"academic new venture*" OR "academic spin-off*" OR "academic spin-out*"
OR "academic spinoff*" OR "academic spinout*" OR "academic new firm*"
OR "academic entrepreneurial" OR "academic entrepreneurship" OR
"academic start up*" OR "academic spinning out" OR "faculty new venture*"
OR "faculty spin-off*" OR "faculty spin-out*" OR "faculty spinoff*" OR
"faculty spinout*" OR "faculty new firm*" OR "faculty entrepreneurial" OR
"faculty entrepreneurship" OR "faculty start up*" OR "faculty spinning out"
OR "research-based new venture*" OR "research-based spin-off*" OR
"research-based spin-out*" OR "research-based spinoff*" OR "research-based
spinout*" OR "research-based new firm*" OR "research-based entrepreneurial"
OR "research-based entrepreneurship" OR "research-based start up*" OR
"research-based spinning out" OR "science-based new venture*" OR "science-
based spin-off*" OR "science-based spin-out*" OR "science-based spinoff*"
OR "science-based spinout*" OR "science-based new firm*" OR "science-
based entrepreneurial" OR "science-based entrepreneurship" OR "science-
based start up*" OR "science-based spinning out" OR "scientist new venture*"
OR "scientist spin-off*" OR "scientist spin-out*" OR "scientist spinoff*" OR
"scientist spinout*" OR "scientist new firm*" OR "scientist entrepreneurial"
OR "scientist entrepreneurship" OR "scientist start up*" OR "scientist spinning
out" OR "university new venture*" OR "university spin-off*" OR "university
spin-out*" OR "university spinoff*" OR "university spinout*" OR "university
new firm*" OR "university entrepreneurial" OR "university entrepreneurship"
OR "university start up*" OR "university spinning out”
2 !
Appendix 2 – Articles from structured literature review
Author(s) Title Citations Journal
The Journal of Technology
Criaco et al. (2013) ‘‘To have and have not’’: founders’ human capital and university start-up survival 10 GS (1 ISI) Transfer
Success factors of university-spin-offs: Regional government support programs
Sternberg (2014) versus regional environment 7 GS Technovation
Founding team composition and early performance of university—Based spin-off
Visintin and Pittino (2014) companies 3 GS Technovation
Empirical analysis of the effect of Japanese university spinoffs' social networks on Technological Forecasting and
Hirai et al. (2013) their performance 2 GS Social Change
Journal of technology management
Bigliardi et al. (2013) Evaluating Performance of University Spin-Off Companies: Lessons from Italy 2 GS & innovation
Value creation from the innovation environment: partnership strategies in
Lubik et al. (2013) university spin-outs 5 GS (3 ISI) R&D Management
Harnessing University Entrepreneurship for Economic Growth: Factors of Success
Hayter (2013) Among University 4 GS (2 ISI) Economic Development Quarterly
Marion et al. (2012) The university entrepreneur: a census and survey of attributes and outcomes 6 GS R&D Management
Fuller and Rothaermel (2012) When Stars Shine: The Effects Of Faculty Founders On New Technology Ventures 9 GS (2 ISI) Strategic Entrepreneurship Journal
Strategic entrepreneurship, resource orchestration and growing spin-offs from Technology Analysis & Strategic
Wright et al. (2012) universities 9 GS (1 ISI) Management
Technology Analysis & Strategic
Zerbinati et al. (2012) Nurture or nature? The growth paradox of research-based spin-offs 1 GS Management
The effectiveness of university knowledge spillovers: Performance differences
Wennberg et al. (2011) between university spinoffs and corporate spinoffs 52 GS (14 ISI) Research Policy
Entrepreneurial Origin, Technological Knowledge, and the Growth of Spin-Off
Clarysse et al. (2011) Companies 54 GS (18 ISI) Journal of Management Studies
The Relationship Between Knowledge Transfer, Top Management Team Entrepreneurship Theory and
Knockaert et al. (2011) Composition, and Performance: The Case of Science-Based Entrepreneurial Firms 48 GS (12 ISI) Practice
3 !
The performance of university spin-offs: an exploratory analysis using venture The Journal of Technology
Zhang (2009) capital data 85 GS (26 ISI) Transfer
Clarysse et al. (2007) Academic spin-offs, formal technology transfer and capital raising 97 GS (34 ISI) Industrial and Corporate Change
Measuring the performance of Oxford University, Oxford Brookes University and
Lawton Smith and Ho (2006) the government laboratories’ spin-off companies 82 GS (21 ISI) Research Policy
The impact of network capabilities and entrepreneurial orientation on university
Walter et al. (2006) spin-off performance 461 GS (116 ISI) Journal of Business Venturing
University start-up formation and technology licensing with firms that go public: a
Powers and McDougall (2005) resource-based view of academic entrepreneurship 348 GS (95 ISI) Journal of Business Venturing
Shane and Stuart (2002) Organizational Endowments and the Performance of University Start-Ups 975 GS (333 ISI) Management science
Valuing University‐ Based Firms: The Effects of Academic Affiliation on IPO Entrepreneurship Theory and
Bonardo et al. (2011) Performance 34 GS Practice
The Journal of Technology
Colombo et al. (2010) Dynamics of science-based entrepreneurship 42 GS (15 ISI) Transfer
Spinouts from academic institutions: a literature review with suggestions for The Journal of Technology
Djokovic and Souitaris (2008) further research 178 GS (42 ISI) Transfer
Entrepreneurial orientation, technology transfer and spinoff performance of US
O'Shea et al. (2005) universities 456 GS Research Policy
Zucker et al. (1998) Geographically localized knowledge: spillovers or markets? 824 GS Economic Inquiry
Di Gregorio and Shane (2003) Why do some universities generate more start-ups than others? 912 GS Research policy
Roberts and Hauptman (1986) The process of technology transfer to the new biomedical and pharmaceutical firm 40 GS (16 ISI) Research Policy
Birley (1986) The role of networks in the entrepreneurial process 1626 GS Journal of business venturing
Gurdon and Samsom (2010) A longitudinal study of success and failure among scientist-started ventures 34 GS Technovation
Mustar et al. (2008) University spin-off firms: lessons from ten years of experience in Europe 54 GS Science and Public Policy
Social Capital Through Networks: The Case Of University Spin‐ Off Firms In Tijdschrift voor economische en
Soetanto and Van Geenhuizen (2010) Different Stages 9 GS (6 ISI) sociale geografie
4 !
Rothaermel and Thursby (2005) Incubator firm failure or graduation?: The role of university linkages 207 GS Research policy
Knowledge sources of entrepreneurship: Firm formation by academic, user and
Agarwal and Shah (2014) employee innovators 2 GS Research Policy
Ortín-Ángel and Vendrell-Herrero University spin-offs vs. other NTBFs: Total factor productivity differences at
(2014) outset and evolution 4 GS Technovation
Academic spin-offs, corporate spin-outs and company internal start-ups as The Journal of Technology
Festel (2013) technology transfer approach 11 GS (2 ISI) Transfer
Rothaermel et al. (2007) University entrepreneurship: a taxonomy of the literature 640 GS (216 ISI) Industrial and corporate change
Assessment of proposals for new technology ventures in the UK: characteristics of
De Coster and Butler (2005) university spin-off companies 78 GS (20 ISI) Technovation
Vohora et al. (2004) Critical junctures in the development of university high-tech spinout companies 606 GS (180 ISI) Research policy
Shane (2004) Academic entrepreneurship: University spinoffs and wealth creation 966 GS
A comparative study of new venture top management team composition, dynamics
Ensley and Hmieleski (2005) and performance between university-based and independent start-ups 214 GS (59 ISI) Research Policy
The business of biotechnology:
Nelsen (1991) The lifeblood of biotechnology: University-industry technology transfer 33 GS from the bench to the street
The transfer and commercialization of university-developed medical imaging IEEE Transactions on Engineering
del Campo et al. (1999) technology: opportunities and problems 28 (9 ISI) Management
Why are some university researchers more likely to create spin-offs than others?
Landry et al. (2006) Evidence from Canadian universities 172 GS Research Policy
Ortín-Ángel and Vendrell-Herrero
(2010) Why do university spin-offs attract more venture capitalists? 8 GS Venture Capital
5

Appendix 3 - Codebook
Variable Coding instructions Variable in analysis Measure

General information

Forny ID Unique ID from FORNY Database Nominal

OrgNr Organizational number of the company Nominal

Initial Name Name of the company Nominal

Comments on coding problems and method of
Comment valuation Nominal

Founded Year Year of founding Scale

TTO Name of TTO involved Nominal

Year of shutdown for unsuccessful firms, acquisition
for acquired firms, or year of last annual report for
Year T companies still operating Ordinal

Age at T Year of founding minus year T Control variable Scale

M&A Year Year of acquisition for acquired companies. Ordinal

Valuation information

Industry(Damodaran) Industry mapping in accordance with Damodaran Nominal

Industry(Ycharts) Industry mapping in accordance with Ycharts Nominal

Coded as either 'unsucessfull', 'Still operating' or
Value Categorical 'M&A' Nominal

Value Integer Unsuccessfull=0, Still operating=1, M&A=2 Ordinal

Value of OSEBX index on Jan 1st, second year of
OSEBX operations Scale

Market Size(Ycharts) Market size of the industry from Ycharts Control Variable Scale

EV/EBITDA (Damodaran) Multiple retrived from Damodaran Scale

EV/EBITDA (Ycharts) Multiple retrived from Ycharts Scale

EV/Revenue (Damodaran) Multiple retrived from Damodaran Scale

EV/Revenue (Ycharts) Multiple retrived from Ycharts Scale

Yearly revenue between founding 15 variables indicating yearly revenue from founding
and year T to year T Scale

Earnings Year T Earnings in year T retrieved from Proff Scale

EBITDA Year T EBITDA in year T retrieved from Proff Scale

Year-over-year Growth Between 13 variables indicating revenue growth year-over-year

year T and founding from founding til year T Scale

Geometric growth in revenues from first year of sales

Geometric Growth til year T Scale

Aritmetic Growth Arithmetic growth in revenues from first year of sales til Year T Scale
Value of company using PEG multiple retrived from
Value (PEG_Damodaran) Damodaran Scale
6

Value of company using PEG multiple retrived from

Value (PEG_Ycharts) Ycharts Scale
Value of company using EV/EBITDA multiple
Value(EV/EBITDA_Damodaran) retrived from Damodaran Scale
Value of company using EV/EBITDA multiple
Value (EV/EBITDA_Ycharts) retrived from Ycharts Scale
Value of company using EV/Sales multiple retrived
Value(EV/Sales_Damodaran) from Damodaran Scale
Value of company using EV/Sales multiple retrived
Value(EV/Sales_Ycharts) from Ycharts Scale

Debt year T Debt on balance sheet in year T - retrieved from Proff. Scale
Cash and cash equivalents on balance sheet in year T -
Cash & Equivalents year T retrieved from Proff. Scale
The value of the company based on the chosen
Value(before discount) multiples. Scale
Coded '1' if the company has conducted their first
Conducted first sale commercial sale Ordinal
Illiquidity discount in accordance with Silber (see
Iliquidity discount appendix 6). Scale
The estimated company value after the illiquidity
Company Value discount is applied. Scale

Company Value Transformed 1+ln(Company Value) Dependent variable Scale

Human Resources
Coded as either 'student', 'Forskningsassistent', 'PhD.
kandidat', 'Amanuensis', 'Førsteamanuensis',
'Professor', 'Departementsleder eller
forskningsinstituttleder', or 'Institutt professor', based
Team Technical Experience on the highest academic rank amongst the founders. Ordinal

Team Technical Coded '1' if the company has technical experience in

Experience(integer) the founding team Ordinal
Coded '1' if the company has marketing experience in
Team Marketing Experience the founding team Ordinal
Coded '1' if the company has financial experience in
Team Financial Experience the founding team Ordinal
Coded '1' if the company has management experience
Team ManagementExperience in the founding team Ordinal
Coded '1' if the company has startup experience in the
Team Startup Experience founding team Ordinal
Coded '1' if the company has industry experience in
Team Industry Experience the founding team Ordinal

The sum of the binary variables 'Team technical

experience(integer)', 'Team marketing experience',
'Team financial experience', 'Team management
experience', 'Team startup experience', and 'Team
Team Heterogeneity industy experience'. Scale

Team Heterogeneity Transformed 1+ln(Team Heterogeneity) Independent variable Scale

Coded '1' if at least one of the technology inventors
Inventor on Team are on the fouding team Independent variable Ordinal
The total citaton count up until 2 years after the
company is founded of all founders as calculate by
Inventor Renownedness Google Scholar. Scale
Inventor Renownedness
Transformed 1+ln(Inventor Renownedness) Independent variable Scale
7

Social Resources

Parent Institution Name of parent university Nominal

Ranking of parent university in accordance with
Parent Ranking Webometrics Scale

Quality of parent university Inverse of parent ranking (1000/parent_rank) Independent variable Scale

The total number of years the founders have

Team Years of Employment at worked/researched at the parent university up until 2
Parent years after the company was founded. Independent variable Scale

Financial Resources
Coded '1' if the company has received investor
Investor Capital funding during its first two years of operations

Technological Resources
The total number of patents filed during or prior to 2
years after the company being founded where the
#Patents Filed company is listed as assignee Scale

#Claims on Patents Filed The total number of citations on all filed patents Scale

Name of patents filed The names on filed patents Nominal

The total number of patents published during or prior

to 2 years after the company being founded where the
#Patents Published company is listed as assignee Scale

# Claims on Patents Published The total number of citations on all published patents Scale

Name of patents published The names on published patents Nominal

Number of patents The sum of #patents filed and #patents published Scale

Number of patents Transformed 1-1/(1+Number of patents) Independent variable Scale

8
Appendix 4 - Descriptives
Descriptives After Transformation
N Minimum Maximum Mean Std. Deviation Variance Skewness Kurtosis Shapiro-Wilk
Variable Statistic Statistic Statistic Statistic Statistic Statistic Statistic Std. Error Statistic Std. Error Statistic df Sig.
Company Value (transformed) 63 0 20,0407519 6,88383237 8,359834129 69,88682667 0,42352599 0,30158857 -1,797988 0,59484062 0,68992287 63 3,0035E-10
Human Resources
Inventor on founding team 63 0 1 0,88888889 0,316793976 0,100358423 -2,5356501 0,30158857 4,57377049 0,59484062 0,36396492 63 4,8484E-15
Inventor renownedness (transformed) 63 0 9,7217858 4,27373482 3,192697293 10,193316 -0,1792133 0,30158857 -1,3975119 0,59484062 0,88828618 63 3,3944E-05
Team heterogeneity (transformed) 63 0 1,94591015 1,05732398 0,411302857 0,16917004 -0,0257955 0,30158857 0,02511675 0,59484062 0,90809878 63 0,00018432
Social Resources
Ties to parent university 63 0 54 17,3968254 13,70959128 187,952893 0,67046479 0,30158857 -0,4408559 0,59484062 0,92285007 63 0,00072782
Parent university quality (transformed) 63 0,1496334 14,7058824 5,98954308 4,867322128 23,6908247 0,58323494 0,30158857 -0,5790366 0,59484062 0,83427262 63 6,621E-07
Financial Resources
Investor capital 63 0 1 0,20634921 0,407934615 0,16641065 1,48689852 0,30158857 0,21672636 0,59484062 0,4968968 63 2,3041E-13
Technological Resources
Number of patents (transformed) 63 0 0,99047619 0,26945744 0,390995877 0,152877776 0,82089374 0,30158857 -1,2697309 0,59484062 0,64416791 63 4,3562E-11
Control Variables
Company age 63 1 13 8,36507937 3,456184673 11,94521249 -0,4935447 0,30158857 -0,8438655 0,59484062 0,93212978 63 0,0018282
Market size (MUSD) 63 25081,0071 1415745,12 429837,388 368827,953 1,36034E+11 1,3648623 0,30158857 1,46986127 0,59484062 0,83320955 63 6,1759E-07
Descriptives Before Transformation
N Minimum Maximum Mean Std. Deviation Variance Skewness Kurtosis Shapiro-Wilk
Variable Statistic Statistic Statistic Statistic Statistic Statistic Statistic Std. Error Statistic Std. Error Statistic df Sig.
Company Value 63 0 505345000 30847241,8 97404974,1 9,48773E+15 4,05987085 0,30158857 16,0063395 0,59484062 0,34808519 63 3,1831E-15
Human Resources
Inventor on founding team 63 0 1 0,88888889 0,316793976 0,100358423 -2,5356501 0,30158857 4,57377049 0,59484062 0,36396492 63 4,8484E-15
Inventor renownedness 63 68 6683 934,333333 1512,818551 2288619,968 2,29378973 0,30158857 5,47323294 0,59484062 0,61269135 63 1,2705E-11
Team heterogeneity 63 0 104 3,41269841 13,24780556 175,5043523 7,29071394 0,30158857 55,9121035 0,59484062 0,2389605 63 2,1285E-16
Social Resources
Ties to parent university 63 0 54 17,3968254 13,70959128 187,952893 0,67046479 0,30158857 -0,4408559 0,59484062 0,92285007 63 0,00072782
Parent university quality (1000/rank) 63 0 6 2,12698413 1,313602084 1,725550435 0,94886563 0,30158857 0,40611251 0,59484062 0,87110329 63 8,8208E-06
Financial Resources
Investor capital 63 0 1 0,20634921 0,407934615 0,16641065 1,48689852 0,30158857 0,21672636 0,59484062 0,4968968 63 2,3041E-13
Technological Resources
Number of patents 63 0 16676 1219,22222 2749,235527 7558295,982 3,80719126 0,30158857 17,061174 0,59484062 0,49515381 63 2,1806E-13
Control Variables
Company age 63 1 13 8,36507937 3,456184673 11,94521249 -0,4935447 0,30158857 -0,8438655 0,59484062 0,93212978 63 0,0018282
Market size (MUSD) 63 25081,0071 1415745,12 429837,388 368827,953 1,36034E+11 1,3648623 0,30158857 1,46986127 0,59484062 0,83320955 63 6,1759E-07
9! !

Appendix 5 - SPSS Model Information

b
Variables Entered/Removed

Variables Variables
Model Entered Removed Method
1 Age_at_T, . Enter
Market
a
Size_Ycharts
2 Team_ . Enter
Heterogeneity
_Transformed
, Inventor_
Renownednes
s_
Transformed,
Investor_
capital,
Quality_Of_
Parent_
Transformed,
Inventor_on_
Team,
Number_of_
patents_
Transformed,
Ties_To_
a
Parent
a. All requested variables entered.

b. Dependent Variable:
Company_Value_Transformed

c
Model Summary

Change Statistics
Adjusted R Std. Error of R Square
Model R R Square Square the Estimate Change F Change df1
a
1 .495 .245 .220 7.38255 .245 9.751 2
b
2 .685 .469 .378 6.59086 .223 3.183 7

a. Predictors: (Constant), Age_at_T, Market Size_Ycharts

b. Predictors: (Constant), Age_at_T, Market Size_Ycharts, Team_Heterogeneity_Transformed,

Inventor_Renownedness_Transformed, Investor_capital, Quality_Of_Parent_Transformed,
Inventor_on_Team, Number_of_patents_Transformed, Ties_To_Parent

c. Dependent Variable: Company_Value_Transformed

c
Model Summary

Change Statistics
Durbin-
Model df2 Sig. F Change Watson
1 60 .000

2 53 .007 1.837
Page 1

c. Dependent Variable: Company_Value_Transformed

10! !

c
ANOVA

Sum of
Model Squares df Mean Square F Sig.
a
1 Regression 1062.859 2 531.430 9.751 .000

Residual 3270.124 60 54.502

Total 4332.983 62
b
2 Regression 2030.691 9 225.632 5.194 .000

Residual 2302.292 53 43.439

Total 4332.983 62

a. Predictors: (Constant), Age_at_T, Market Size_Ycharts

b. Predictors: (Constant), Age_at_T, Market Size_Ycharts,

Team_Heterogeneity_Transformed, Inventor_Renownedness_Transformed,
Investor_capital, Quality_Of_Parent_Transformed, Inventor_on_Team,
Number_of_patents_Transformed, Ties_To_Parent

c. Dependent Variable: Company_Value_Transformed

b
Excluded Variables

Partial
Model Beta In t Sig. Correlation
a
1 Inventor_on_Team .122 1.054 .296 .136
a
Team_Heterogeneity_ .231 2.095 .040 .263
Transformed
a
Number_of_patents_ .257 2.223 .030 .278
Transformed
a
Quality_Of_Parent_ .268 2.288 .026 .286
Transformed
a
Investor_capital -.077 -.644 .522 -.084
a
Inventor_Renownedness_ .164 1.428 .158 .183
Transformed
a
Ties_To_Parent .343 3.098 .003 .374

a. Predictors in the Model: (Constant), Age_at_T, Market Size_Ycharts

b. Dependent Variable: Company_Value_Transformed

b
Excluded Variables

Collinearity Statistics
Minimum
Model Tolerance VIF Tolerance
1 Inventor_on_Team .943 1.060 .943

Team_Heterogeneity_ .979 1.021 .979

Transformed
Number_of_patents_ .881 1.135 .881
Transformed
Quality_Of_Parent_ .860 1.163 .860 Page 1
Transformed
Investor_capital .888 1.126 .887

Inventor_Renownedness_ .938 1.066 .938

Transformed
Ties_To_Parent .896 1.116 .896

b. Dependent Variable: Company_Value_Transformed

 Correlations
11! !
12! !
13! !
14! !
15! !

a
Residuals Statistics

Minimum Maximum Mean Std. Deviation N

Predicted Value -4.6374 22.9894 6.8838 5.72303 63

Std. Predicted Value -2.013 2.814 .000 1.000 63

Standard Error of 1.645 3.875 2.570 .545 63

Predicted Value
Adjusted Predicted Value -5.2705 26.3584 6.9255 5.99037 63

Residual -13.28685 13.41070 .00000 6.09375 63

Std. Residual -2.016 2.035 .000 .925 63

Stud. Residual -2.157 2.265 -.003 1.017 63

Deleted Residual -15.21210 16.86695 -.04166 7.39411 63

Stud. Deleted Residual -2.237 2.360 .000 1.033 63

Mahal. Distance 2.878 20.452 8.857 4.287 63

Cook's Distance .000 .176 .022 .035 63

Centered Leverage Value .046 .330 .143 .069 63

a. Dependent Variable: Company_Value_Transformed

Page 1
 16! !

Appendix 6 - Normality of Residuals

Histogram

Dependent Variable: Company_Value_Transformed

Mean = 3.23E-16
12 Std. Dev. = 0.925
N = 63

10

8
Frequency

0
-3 -2 -1 0 1 2 3

Regression Standardized Residual

Tests of Normality
a
Kolmogorov-Smirnov Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
*
Standardized Residual .072 63 .200 .980 63 .395

a. Lilliefors Significance Correction

*. This is a lower bound of the true significance.

Page 1
17! !

Appendix 7 - Illiquidity Discount

Silber’s (1991) equation:

LN(RPRS)=4,33+ 0,036*LN(REV) - 0,142*LN (RBRT) + 0,174*(DERN) +

0,332*DCUST

RPRS: Relative price of restricted stock expressed in percentage terms

[(pr/pc)*100]
REV: Firms revenues in MUSD
RBRT: Size of the restricted block of shares relative to total common stock (in per
cent, and equals 100 in our case of private firms).
DERN: Dummy variable equal to 1 if the firm has positive earnings.
DCUST: Dummy variable equal to 1 if there is a customer relationship between
investor and firm.

Calculated illiquidity discount in valuation model: